tts models support (#2033)

Co-authored-by: luowei <glpat-EjySCyNjWiLqAED-YmwM>
Co-authored-by: crazywoola <427733928@qq.com>
Co-authored-by: crazywoola <100913391+crazywoola@users.noreply.github.com>
Co-authored-by: Yeuoly <45712896+Yeuoly@users.noreply.github.com>

api/Dockerfile

@@ -30,7 +30,7 @@ ENV TZ UTC
 WORKDIR /app/api
 
 RUN apt-get update \
-    && apt-get install -y --no-install-recommends bash curl wget vim nodejs \
+    && apt-get install -y --no-install-recommends bash curl wget vim nodejs ffmpeg \
     && apt-get autoremove \
     && rm -rf /var/lib/apt/lists/*
 

api/app.py

@@ -124,6 +124,7 @@ def load_user_from_request(request_from_flask_login):
     else:
         return None
 
+
 @login_manager.unauthorized_handler
 def unauthorized_handler():
     """Handle unauthorized requests."""

api/controllers/console/app/audio.py

@@ -32,9 +32,10 @@ class ChatMessageAudioApi(Resource):
         file = request.files['file']
 
         try:
-            response = AudioService.transcript(
+            response = AudioService.transcript_asr(
                 tenant_id=app_model.tenant_id,
                 file=file,
+                promot=app_model.app_model_config.pre_prompt
             )
 
             return response
@@ -64,4 +65,46 @@ class ChatMessageAudioApi(Resource):
             raise InternalServerError()
 
 
+class ChatMessageTextApi(Resource):
+    @setup_required
+    @login_required
+    @account_initialization_required
+    def post(self, app_id):
+        app_id = str(app_id)
+        app_model = _get_app(app_id, None)
+        try:
+            response = AudioService.transcript_tts(
+                tenant_id=app_model.tenant_id,
+                text=request.form['text'],
+                streaming=False
+            )
+
+            return {'data': response.data.decode('latin1')}
+        except services.errors.app_model_config.AppModelConfigBrokenError:
+            logging.exception("App model config broken.")
+            raise AppUnavailableError()
+        except NoAudioUploadedServiceError:
+            raise NoAudioUploadedError()
+        except AudioTooLargeServiceError as e:
+            raise AudioTooLargeError(str(e))
+        except UnsupportedAudioTypeServiceError:
+            raise UnsupportedAudioTypeError()
+        except ProviderNotSupportSpeechToTextServiceError:
+            raise ProviderNotSupportSpeechToTextError()
+        except ProviderTokenNotInitError as ex:
+            raise ProviderNotInitializeError(ex.description)
+        except QuotaExceededError:
+            raise ProviderQuotaExceededError()
+        except ModelCurrentlyNotSupportError:
+            raise ProviderModelCurrentlyNotSupportError()
+        except InvokeError as e:
+            raise CompletionRequestError(e.description)
+        except ValueError as e:
+            raise e
+        except Exception as e:
+            logging.exception("internal server error.")
+            raise InternalServerError()
+
+
 api.add_resource(ChatMessageAudioApi, '/apps/<uuid:app_id>/audio-to-text')
+api.add_resource(ChatMessageTextApi, '/apps/<uuid:app_id>/text-to-audio')

api/controllers/console/explore/audio.py

@@ -29,7 +29,7 @@ class ChatAudioApi(InstalledAppResource):
         file = request.files['file']
 
         try:
-            response = AudioService.transcript(
+            response = AudioService.transcript_asr(
                 tenant_id=app_model.tenant_id,
                 file=file,
             )
@@ -61,4 +61,46 @@ class ChatAudioApi(InstalledAppResource):
             raise InternalServerError()
 
 
+class ChatTextApi(InstalledAppResource):
+    def post(self, installed_app):
+        app_model = installed_app.app
+        app_model_config: AppModelConfig = app_model.app_model_config
+
+        if not app_model_config.text_to_speech_dict['enabled']:
+            raise AppUnavailableError()
+
+        try:
+            response = AudioService.transcript_tts(
+                tenant_id=app_model.tenant_id,
+                text=request.form['text'],
+                streaming=False
+            )
+            return {'data': response.data.decode('latin1')}
+        except services.errors.app_model_config.AppModelConfigBrokenError:
+            logging.exception("App model config broken.")
+            raise AppUnavailableError()
+        except NoAudioUploadedServiceError:
+            raise NoAudioUploadedError()
+        except AudioTooLargeServiceError as e:
+            raise AudioTooLargeError(str(e))
+        except UnsupportedAudioTypeServiceError:
+            raise UnsupportedAudioTypeError()
+        except ProviderNotSupportSpeechToTextServiceError:
+            raise ProviderNotSupportSpeechToTextError()
+        except ProviderTokenNotInitError as ex:
+            raise ProviderNotInitializeError(ex.description)
+        except QuotaExceededError:
+            raise ProviderQuotaExceededError()
+        except ModelCurrentlyNotSupportError:
+            raise ProviderModelCurrentlyNotSupportError()
+        except InvokeError as e:
+            raise CompletionRequestError(e.description)
+        except ValueError as e:
+            raise e
+        except Exception as e:
+            logging.exception("internal server error.")
+            raise InternalServerError()
+
+
 api.add_resource(ChatAudioApi, '/installed-apps/<uuid:installed_app_id>/audio-to-text', endpoint='installed_app_audio')
+api.add_resource(ChatTextApi, '/installed-apps/<uuid:installed_app_id>/text-to-audio', endpoint='installed_app_text')

api/controllers/console/explore/parameter.py

@@ -31,6 +31,7 @@ class AppParameterApi(InstalledAppResource):
         'suggested_questions': fields.Raw,
         'suggested_questions_after_answer': fields.Raw,
         'speech_to_text': fields.Raw,
+        'text_to_speech': fields.Raw,
         'retriever_resource': fields.Raw,
         'annotation_reply': fields.Raw,
         'more_like_this': fields.Raw,
@@ -51,6 +52,7 @@ class AppParameterApi(InstalledAppResource):
             'suggested_questions': app_model_config.suggested_questions_list,
             'suggested_questions_after_answer': app_model_config.suggested_questions_after_answer_dict,
             'speech_to_text': app_model_config.speech_to_text_dict,
+            'text_to_speech': app_model_config.text_to_speech_dict,
             'retriever_resource': app_model_config.retriever_resource_dict,
             'annotation_reply': app_model_config.annotation_reply_dict,
             'more_like_this': app_model_config.more_like_this_dict,

api/controllers/service_api/app/app.py

@@ -33,6 +33,7 @@ class AppParameterApi(AppApiResource):
         'suggested_questions': fields.Raw,
         'suggested_questions_after_answer': fields.Raw,
         'speech_to_text': fields.Raw,
+        'text_to_speech': fields.Raw,
         'retriever_resource': fields.Raw,
         'annotation_reply': fields.Raw,
         'more_like_this': fields.Raw,
@@ -52,6 +53,7 @@ class AppParameterApi(AppApiResource):
             'suggested_questions': app_model_config.suggested_questions_list,
             'suggested_questions_after_answer': app_model_config.suggested_questions_after_answer_dict,
             'speech_to_text': app_model_config.speech_to_text_dict,
+            'text_to_speech': app_model_config.text_to_speech_dict,
             'retriever_resource': app_model_config.retriever_resource_dict,
             'annotation_reply': app_model_config.annotation_reply_dict,
             'more_like_this': app_model_config.more_like_this_dict,

api/controllers/service_api/app/audio.py

@@ -10,6 +10,7 @@ from controllers.service_api.wraps import AppApiResource
 from core.errors.error import ModelCurrentlyNotSupportError, ProviderTokenNotInitError, QuotaExceededError
 from core.model_runtime.errors.invoke import InvokeError
 from flask import request
+from flask_restful import reqparse
 from models.model import App, AppModelConfig
 from services.audio_service import AudioService
 from services.errors.audio import (AudioTooLargeServiceError, NoAudioUploadedServiceError,
@@ -27,9 +28,10 @@ class AudioApi(AppApiResource):
         file = request.files['file']
 
         try:
-            response = AudioService.transcript(
+            response = AudioService.transcript_asr(
                 tenant_id=app_model.tenant_id,
                 file=file,
+                end_user=end_user
             )
 
             return response
@@ -58,4 +60,48 @@ class AudioApi(AppApiResource):
             logging.exception("internal server error.")
             raise InternalServerError()
 
+
+class TextApi(AppApiResource):
+    def post(self, app_model: App, end_user):
+        parser = reqparse.RequestParser()
+        parser.add_argument('text', type=str, required=True, nullable=False, location='json')
+        parser.add_argument('user', type=str, required=True, nullable=False, location='json')
+        args = parser.parse_args()
+
+        try:
+            response = AudioService.transcript_tts(
+                tenant_id=app_model.tenant_id,
+                text=args['text'],
+                end_user=args['user'],
+                streaming=False
+            )
+
+            return response
+        except services.errors.app_model_config.AppModelConfigBrokenError:
+            logging.exception("App model config broken.")
+            raise AppUnavailableError()
+        except NoAudioUploadedServiceError:
+            raise NoAudioUploadedError()
+        except AudioTooLargeServiceError as e:
+            raise AudioTooLargeError(str(e))
+        except UnsupportedAudioTypeServiceError:
+            raise UnsupportedAudioTypeError()
+        except ProviderNotSupportSpeechToTextServiceError:
+            raise ProviderNotSupportSpeechToTextError()
+        except ProviderTokenNotInitError as ex:
+            raise ProviderNotInitializeError(ex.description)
+        except QuotaExceededError:
+            raise ProviderQuotaExceededError()
+        except ModelCurrentlyNotSupportError:
+            raise ProviderModelCurrentlyNotSupportError()
+        except InvokeError as e:
+            raise CompletionRequestError(e.description)
+        except ValueError as e:
+            raise e
+        except Exception as e:
+            logging.exception("internal server error.")
+            raise InternalServerError()
+
+
 api.add_resource(AudioApi, '/audio-to-text')
+api.add_resource(TextApi, '/text-to-audio')

api/controllers/web/app.py

@@ -32,6 +32,7 @@ class AppParameterApi(WebApiResource):
         'suggested_questions': fields.Raw,
         'suggested_questions_after_answer': fields.Raw,
         'speech_to_text': fields.Raw,
+        'text_to_speech': fields.Raw,
         'retriever_resource': fields.Raw,
         'annotation_reply': fields.Raw,
         'more_like_this': fields.Raw,
@@ -51,6 +52,7 @@ class AppParameterApi(WebApiResource):
             'suggested_questions': app_model_config.suggested_questions_list,
             'suggested_questions_after_answer': app_model_config.suggested_questions_after_answer_dict,
             'speech_to_text': app_model_config.speech_to_text_dict,
+            'text_to_speech': app_model_config.text_to_speech_dict,
             'retriever_resource': app_model_config.retriever_resource_dict,
             'annotation_reply': app_model_config.annotation_reply_dict,
             'more_like_this': app_model_config.more_like_this_dict,

api/controllers/web/audio.py

@@ -28,7 +28,7 @@ class AudioApi(WebApiResource):
         file = request.files['file']
 
         try:
-            response = AudioService.transcript(
+            response = AudioService.transcript_asr(
                 tenant_id=app_model.tenant_id,
                 file=file,
             )
@@ -59,4 +59,43 @@ class AudioApi(WebApiResource):
             logging.exception("internal server error.")
             raise InternalServerError()
 
+
+class TextApi(WebApiResource):
+    def post(self, app_model: App, end_user):
+        try:
+            response = AudioService.transcript_tts(
+                tenant_id=app_model.tenant_id,
+                text=request.form['text'],
+                end_user=end_user.external_user_id,
+                streaming=False
+            )
+
+            return {'data': response.data.decode('latin1')}
+        except services.errors.app_model_config.AppModelConfigBrokenError:
+            logging.exception("App model config broken.")
+            raise AppUnavailableError()
+        except NoAudioUploadedServiceError:
+            raise NoAudioUploadedError()
+        except AudioTooLargeServiceError as e:
+            raise AudioTooLargeError(str(e))
+        except UnsupportedAudioTypeServiceError:
+            raise UnsupportedAudioTypeError()
+        except ProviderNotSupportSpeechToTextServiceError:
+            raise ProviderNotSupportSpeechToTextError()
+        except ProviderTokenNotInitError as ex:
+            raise ProviderNotInitializeError(ex.description)
+        except QuotaExceededError:
+            raise ProviderQuotaExceededError()
+        except ModelCurrentlyNotSupportError:
+            raise ProviderModelCurrentlyNotSupportError()
+        except InvokeError as e:
+            raise CompletionRequestError(e.description)
+        except ValueError as e:
+            raise e
+        except Exception as e:
+            logging.exception("internal server error.")
+            raise InternalServerError()
+
+
 api.add_resource(AudioApi, '/audio-to-text')
+api.add_resource(TextApi, '/text-to-audio')

api/core/application_manager.py

@@ -555,6 +555,12 @@ class ApplicationManager:
             if 'enabled' in speech_to_text_dict and speech_to_text_dict['enabled']:
                 properties['speech_to_text'] = True
 
+        # text to speech
+        text_to_speech_dict = copy_app_model_config_dict.get('text_to_speech')
+        if text_to_speech_dict:
+            if 'enabled' in text_to_speech_dict and text_to_speech_dict['enabled']:
+                properties['text_to_speech'] = True
+
         # sensitive word avoidance
         sensitive_word_avoidance_dict = copy_app_model_config_dict.get('sensitive_word_avoidance')
         if sensitive_word_avoidance_dict:

api/core/entities/application_entities.py

@@ -219,6 +219,7 @@ class AppOrchestrationConfigEntity(BaseModel):
     show_retrieve_source: bool = False
     more_like_this: bool = False
     speech_to_text: bool = False
+    text_to_speech: bool = False
     sensitive_word_avoidance: Optional[SensitiveWordAvoidanceEntity] = None
 
 
@@ -283,7 +284,6 @@ class ApplicationGenerateEntity(BaseModel):
     query: Optional[str] = None
     files: list[FileObj] = []
     user_id: str
-
     # extras
     stream: bool
     invoke_from: InvokeFrom

api/core/model_manager.py

@@ -12,6 +12,7 @@ from core.model_runtime.model_providers.__base.large_language_model import Large
 from core.model_runtime.model_providers.__base.moderation_model import ModerationModel
 from core.model_runtime.model_providers.__base.rerank_model import RerankModel
 from core.model_runtime.model_providers.__base.speech2text_model import Speech2TextModel
+from core.model_runtime.model_providers.__base.tts_model import TTSModel
 from core.model_runtime.model_providers.__base.text_embedding_model import TextEmbeddingModel
 from core.provider_manager import ProviderManager
 
@@ -144,7 +145,7 @@ class ModelInstance:
             user=user
         )
 
-    def invoke_speech2text(self, file: IO[bytes], user: Optional[str] = None, **params) \
+    def invoke_speech2text(self, file: IO[bytes], user: Optional[str] = None) \
             -> str:
         """
         Invoke large language model
@@ -161,8 +162,29 @@ class ModelInstance:
             model=self.model,
             credentials=self.credentials,
             file=file,
+            user=user
+        )
+
+    def invoke_tts(self, content_text: str, streaming: bool, user: Optional[str] = None) \
+            -> str:
+        """
+        Invoke large language model
+
+        :param content_text: text content to be translated
+        :param user: unique user id
+        :param streaming: output is streaming
+        :return: text for given audio file
+        """
+        if not isinstance(self.model_type_instance, TTSModel):
+            raise Exception(f"Model type instance is not TTSModel")
+
+        self.model_type_instance = cast(TTSModel, self.model_type_instance)
+        return self.model_type_instance.invoke(
+            model=self.model,
+            credentials=self.credentials,
+            content_text=content_text,
             user=user,
-            **params
+            streaming=streaming
         )
 
 

api/core/model_runtime/entities/model_entities.py

@@ -15,7 +15,7 @@ class ModelType(Enum):
     RERANK = "rerank"
     SPEECH2TEXT = "speech2text"
     MODERATION = "moderation"
-    # TTS = "tts"
+    TTS = "tts"
     # TEXT2IMG = "text2img"
 
     @classmethod
@@ -33,6 +33,8 @@ class ModelType(Enum):
             return cls.RERANK
         elif origin_model_type == 'speech2text' or origin_model_type == cls.SPEECH2TEXT.value:
             return cls.SPEECH2TEXT
+        elif origin_model_type == 'tts' or origin_model_type == cls.TTS.value:
+            return cls.TTS
         elif origin_model_type == cls.MODERATION.value:
             return cls.MODERATION
         else:
@@ -52,6 +54,8 @@ class ModelType(Enum):
             return 'reranking'
         elif self == self.SPEECH2TEXT:
             return 'speech2text'
+        elif self == self.TTS:
+            return 'tts'
         elif self == self.MODERATION:
             return 'moderation'
         else:
@@ -120,6 +124,10 @@ class ModelPropertyKey(Enum):
     FILE_UPLOAD_LIMIT = "file_upload_limit"
     SUPPORTED_FILE_EXTENSIONS = "supported_file_extensions"
     MAX_CHARACTERS_PER_CHUNK = "max_characters_per_chunk"
+    DEFAULT_VOICE = "default_voice"
+    WORD_LIMIT = "word_limit"
+    AUDOI_TYPE = "audio_type"
+    MAX_WORKERS = "max_workers"
 
 
 class ProviderModel(BaseModel):

api/core/model_runtime/model_providers/__base/tts_model.py

+from abc import abstractmethod
+from typing import Optional
+
+from core.model_runtime.entities.model_entities import ModelType
+from core.model_runtime.model_providers.__base.ai_model import AIModel
+
+
+class TTSModel(AIModel):
+    """
+    Model class for ttstext model.
+    """
+    model_type: ModelType = ModelType.TTS
+
+    def invoke(self, model: str, credentials: dict, content_text: str, streaming: bool, user: Optional[str] = None):
+        """
+        Invoke large language model
+
+        :param model: model name
+        :param credentials: model credentials
+        :param content_text: text content to be translated
+        :param streaming: output is streaming
+        :param user: unique user id
+        :return: translated audio file
+        """
+        try:
+            return self._invoke(model=model, credentials=credentials, user=user, streaming=streaming, content_text=content_text)
+        except Exception as e:
+            raise self._transform_invoke_error(e)
+
+    @abstractmethod
+    def _invoke(self, model: str, credentials: dict, content_text: str, streaming: bool, user: Optional[str] = None):
+        """
+        Invoke large language model
+
+        :param model: model name
+        :param credentials: model credentials
+        :param content_text: text content to be translated
+        :param streaming: output is streaming
+        :param user: unique user id
+        :return: translated audio file
+        """
+        raise NotImplementedError

api/core/model_runtime/model_providers/openai/openai.yaml

@@ -20,6 +20,7 @@ supported_model_types:
   - text-embedding
   - speech2text
   - moderation
+  - tts
 configurate_methods:
   - predefined-model
   - customizable-model

api/core/model_runtime/model_providers/openai/tts/tts-1-hd.yaml

+model: tts-1-hd
+model_type: tts
+model_properties:
+  default_voice: 'alloy'
+  word_limit: 120
+  audio_type: 'mp3'
+  max_workers: 5

api/core/model_runtime/model_providers/openai/tts/tts-1.yaml

+model: tts-1
+model_type: tts
+model_properties:
+  default_voice: 'alloy'
+  word_limit: 120
+  audio_type: 'mp3'
+  max_workers: 5

api/core/model_runtime/model_providers/openai/tts/tts.py

+import uuid
+import hashlib
+import subprocess
+from io import BytesIO
+from typing import Optional
+from functools import reduce
+from pydub import AudioSegment
+
+from core.model_runtime.entities.model_entities import ModelPropertyKey
+from core.model_runtime.errors.validate import CredentialsValidateFailedError
+from core.model_runtime.errors.invoke import InvokeBadRequestError
+from core.model_runtime.model_providers.__base.tts_model import TTSModel
+from core.model_runtime.model_providers.openai._common import _CommonOpenAI
+
+from typing_extensions import Literal
+from flask import Response, stream_with_context
+from openai import OpenAI
+import concurrent.futures
+
+
+class OpenAIText2SpeechModel(_CommonOpenAI, TTSModel):
+    """
+    Model class for OpenAI Speech to text model.
+    """
+    def _invoke(self, model: str, credentials: dict, content_text: str, streaming: bool, user: Optional[str] = None) -> any:
+        """
+        _invoke text2speech model
+
+        :param model: model name
+        :param credentials: model credentials
+        :param content_text: text content to be translated
+        :param streaming: output is streaming
+        :param user: unique user id
+        :return: text translated to audio file
+        """
+        self._is_ffmpeg_installed()
+        audio_type = self._get_model_audio_type(model, credentials)
+        if streaming:
+            return Response(stream_with_context(self._tts_invoke_streaming(model=model,
+                                                                           credentials=credentials,
+                                                                           content_text=content_text,
+                                                                           user=user)),
+                            status=200, mimetype=f'audio/{audio_type}')
+        else:
+            return self._tts_invoke(model=model, credentials=credentials, content_text=content_text, user=user)
+
+    def validate_credentials(self, model: str, credentials: dict, user: Optional[str] = None) -> None:
+        """
+        validate credentials text2speech model
+
+        :param model: model name
+        :param credentials: model credentials
+        :param user: unique user id
+        :return: text translated to audio file
+        """
+        try:
+            self._tts_invoke(
+                model=model,
+                credentials=credentials,
+                content_text='Hello world!',
+                user=user
+            )
+        except Exception as ex:
+            raise CredentialsValidateFailedError(str(ex))
+
+    def _tts_invoke(self, model: str, credentials: dict, content_text: str, user: Optional[str] = None) -> any:
+        """
+        _tts_invoke text2speech model
+
+        :param model: model name
+        :param credentials: model credentials
+        :param content_text: text content to be translated
+        :param user: unique user id
+        :return: text translated to audio file
+        """
+        audio_type = self._get_model_audio_type(model, credentials)
+        word_limit = self._get_model_word_limit(model, credentials)
+        max_workers = self._get_model_workers_limit(model, credentials)
+
+        try:
+            sentences = list(self._split_text_into_sentences(text=content_text, limit=word_limit))
+            audio_bytes_list = list()
+
+            # Create a thread pool and map the function to the list of sentences
+            with concurrent.futures.ThreadPoolExecutor(max_workers=max_workers) as executor:
+                futures = [executor.submit(self._process_sentence, sentence, model, credentials) for sentence
+                           in sentences]
+                for future in futures:
+                    try:
+                        audio_bytes_list.append(future.result())
+                    except Exception as ex:
+                        raise InvokeBadRequestError(str(ex))
+
+            audio_segments = [AudioSegment.from_file(BytesIO(audio_bytes), format=audio_type) for audio_bytes in
+                              audio_bytes_list if audio_bytes]
+            combined_segment = reduce(lambda x, y: x + y, audio_segments)
+            buffer: BytesIO = BytesIO()
+            combined_segment.export(buffer, format=audio_type)
+            buffer.seek(0)
+            return Response(buffer.read(), status=200, mimetype=f"audio/{audio_type}")
+        except Exception as ex:
+            raise InvokeBadRequestError(str(ex))
+
+    # Todo: To improve the streaming function
+    def _tts_invoke_streaming(self, model: str, credentials: dict, content_text: str, user: Optional[str] = None) -> any:
+        """
+        _tts_invoke_streaming text2speech model
+
+        :param model: model name
+        :param credentials: model credentials
+        :param content_text: text content to be translated
+        :param user: unique user id
+        :return: text translated to audio file
+        """
+        # transform credentials to kwargs for model instance
+        credentials_kwargs = self._to_credential_kwargs(credentials)
+        voice_name = self._get_model_voice(model, credentials)
+        word_limit = self._get_model_word_limit(model, credentials)
+        audio_type = self._get_model_audio_type(model, credentials)
+        tts_file_id = self._get_file_name(content_text)
+        file_path = f'storage/generate_files/{audio_type}/{tts_file_id}.{audio_type}'
+        try:
+            client = OpenAI(**credentials_kwargs)
+            sentences = list(self._split_text_into_sentences(text=content_text, limit=word_limit))
+            for sentence in sentences:
+                response = client.audio.speech.create(model=model, voice=voice_name, input=sentence.strip())
+                response.stream_to_file(file_path)
+        except Exception as ex:
+            raise InvokeBadRequestError(str(ex))
+
+    def _get_model_voice(self, model: str, credentials: dict) -> Literal["alloy", "echo", "fable", "onyx", "nova", "shimmer"]:
+        """
+        Get voice for given tts model
+
+        :param model: model name
+        :param credentials: model credentials
+        :return: voice
+        """
+        model_schema = self.get_model_schema(model, credentials)
+
+        if model_schema and ModelPropertyKey.DEFAULT_VOICE in model_schema.model_properties:
+            return model_schema.model_properties[ModelPropertyKey.DEFAULT_VOICE]
+
+    def _get_model_audio_type(self, model: str, credentials: dict) -> str:
+        """
+        Get audio type for given tts model
+
+        :param model: model name
+        :param credentials: model credentials
+        :return: voice
+        """
+        model_schema = self.get_model_schema(model, credentials)
+
+        if model_schema and ModelPropertyKey.AUDOI_TYPE in model_schema.model_properties:
+            return model_schema.model_properties[ModelPropertyKey.AUDOI_TYPE]
+
+    def _get_model_word_limit(self, model: str, credentials: dict) -> int:
+        """
+        Get audio type for given tts model
+        :return: audio type
+        """
+        model_schema = self.get_model_schema(model, credentials)
+
+        if model_schema and ModelPropertyKey.WORD_LIMIT in model_schema.model_properties:
+            return model_schema.model_properties[ModelPropertyKey.WORD_LIMIT]
+
+    def _get_model_workers_limit(self, model: str, credentials: dict) -> int:
+        """
+        Get audio max workers for given tts model
+        :return: audio type
+        """
+        model_schema = self.get_model_schema(model, credentials)
+
+        if model_schema and ModelPropertyKey.MAX_WORKERS in model_schema.model_properties:
+            return model_schema.model_properties[ModelPropertyKey.MAX_WORKERS]
+
+    @staticmethod
+    def _split_text_into_sentences(text: str, limit: int, delimiters=None):
+        if delimiters is None:
+            delimiters = set('。！？；\n')
+
+        buf = []
+        word_count = 0
+        for char in text:
+            buf.append(char)
+            if char in delimiters:
+                if word_count >= limit:
+                    yield ''.join(buf)
+                    buf = []
+                    word_count = 0
+                else:
+                    word_count += 1
+            else:
+                word_count += 1
+
+        if buf:
+            yield ''.join(buf)
+
+    @staticmethod
+    def _get_file_name(file_content: str) -> str:
+        hash_object = hashlib.sha256(file_content.encode())
+        hex_digest = hash_object.hexdigest()
+
+        namespace_uuid = uuid.UUID('a5da6ef9-b303-596f-8e88-bf8fa40f4b31')
+        unique_uuid = uuid.uuid5(namespace_uuid, hex_digest)
+        return str(unique_uuid)
+
+    def _process_sentence(self, sentence: str, model: str, credentials: dict):
+        """
+        _tts_invoke openai text2speech model api
+
+        :param model: model name
+        :param credentials: model credentials
+        :param sentence: text content to be translated
+        :return: text translated to audio file
+        """
+        # transform credentials to kwargs for model instance
+        credentials_kwargs = self._to_credential_kwargs(credentials)
+        voice_name = self._get_model_voice(model, credentials)
+
+        client = OpenAI(**credentials_kwargs)
+        response = client.audio.speech.create(model=model, voice=voice_name, input=sentence.strip())
+        if isinstance(response.read(), bytes):
+            return response.read()
+
+    @staticmethod
+    def _is_ffmpeg_installed():
+        try:
+            output = subprocess.check_output("ffmpeg -version", shell=True)
+            if "ffmpeg version" in output.decode("utf-8"):
+                return True
+            else:
+                raise InvokeBadRequestError("ffmpeg is not installed")
+        except Exception:
+            raise InvokeBadRequestError("ffmpeg is not installed")

api/fields/app_fields.py

@@ -19,6 +19,7 @@ model_config_fields = {
     'suggested_questions': fields.Raw(attribute='suggested_questions_list'),
     'suggested_questions_after_answer': fields.Raw(attribute='suggested_questions_after_answer_dict'),
     'speech_to_text': fields.Raw(attribute='speech_to_text_dict'),
+    'text_to_speech': fields.Raw(attribute='text_to_speech_dict'),
     'retriever_resource': fields.Raw(attribute='retriever_resource_dict'),
     'annotation_reply': fields.Raw(attribute='annotation_reply_dict'),
     'more_like_this': fields.Raw(attribute='more_like_this_dict'),

api/migrations/versions/b24be59fbb04_.py

+"""empty message
+
+Revision ID: b24be59fbb04
+Revises: 187385f442fc
+Create Date: 2024-01-17 01:31:12.670556
+
+"""
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = 'b24be59fbb04'
+down_revision = 'de95f5c77138'
+branch_labels = None
+depends_on = None
+
+
+def upgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    with op.batch_alter_table('app_model_configs', schema=None) as batch_op:
+        batch_op.add_column(sa.Column('text_to_speech', sa.Text(), nullable=True))
+
+    # ### end Alembic commands ###
+
+
+def downgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    with op.batch_alter_table('app_model_configs', schema=None) as batch_op:
+        batch_op.drop_column('text_to_speech')
+
+    # ### end Alembic commands ###

api/models/model.py

@@ -146,6 +146,7 @@ class AppModelConfig(db.Model):
     suggested_questions = db.Column(db.Text)
     suggested_questions_after_answer = db.Column(db.Text)
     speech_to_text = db.Column(db.Text)
+    text_to_speech = db.Column(db.Text)
     more_like_this = db.Column(db.Text)
     model = db.Column(db.Text)
     user_input_form = db.Column(db.Text)
@@ -184,6 +185,11 @@ class AppModelConfig(db.Model):
         return json.loads(self.speech_to_text) if self.speech_to_text \
             else {"enabled": False}
 
+    @property
+    def text_to_speech_dict(self) -> dict:
+        return json.loads(self.text_to_speech) if self.text_to_speech \
+            else {"enabled": False}
+
     @property
     def retriever_resource_dict(self) -> dict:
         return json.loads(self.retriever_resource) if self.retriever_resource \
@@ -263,6 +269,7 @@ class AppModelConfig(db.Model):
             "suggested_questions": self.suggested_questions_list,
             "suggested_questions_after_answer": self.suggested_questions_after_answer_dict,
             "speech_to_text": self.speech_to_text_dict,
+            "text_to_speech": self.text_to_speech_dict,
             "retriever_resource": self.retriever_resource_dict,
             "annotation_reply": self.annotation_reply_dict,
             "more_like_this": self.more_like_this_dict,
@@ -289,6 +296,8 @@ class AppModelConfig(db.Model):
         self.suggested_questions_after_answer = json.dumps(model_config['suggested_questions_after_answer'])
         self.speech_to_text = json.dumps(model_config['speech_to_text']) \
             if model_config.get('speech_to_text') else None
+        self.text_to_speech = json.dumps(model_config['text_to_speech']) \
+            if model_config.get('text_to_speech') else None
         self.more_like_this = json.dumps(model_config['more_like_this'])
         self.sensitive_word_avoidance = json.dumps(model_config['sensitive_word_avoidance']) \
             if model_config.get('sensitive_word_avoidance') else None
@@ -323,6 +332,7 @@ class AppModelConfig(db.Model):
             suggested_questions=self.suggested_questions,
             suggested_questions_after_answer=self.suggested_questions_after_answer,
             speech_to_text=self.speech_to_text,
+            text_to_speech=self.text_to_speech,
             more_like_this=self.more_like_this,
             sensitive_word_avoidance=self.sensitive_word_avoidance,
             external_data_tools=self.external_data_tools,

api/requirements.txt

@@ -65,3 +65,4 @@ httpx[socks]~=0.24.1
 pydub~=0.25.1
 matplotlib~=3.8.2
 yfinance~=0.2.35
+pydub~=0.25.1
\ No newline at end of file

api/services/app_model_config_service.py

@@ -95,6 +95,21 @@ class AppModelConfigService:
         if not isinstance(config["speech_to_text"]["enabled"], bool):
             raise ValueError("enabled in speech_to_text must be of boolean type")
 
+        # text_to_speech
+        if 'text_to_speech' not in config or not config["text_to_speech"]:
+            config["text_to_speech"] = {
+                "enabled": False
+            }
+
+        if not isinstance(config["text_to_speech"], dict):
+            raise ValueError("text_to_speech must be of dict type")
+
+        if "enabled" not in config["text_to_speech"] or not config["text_to_speech"]["enabled"]:
+            config["text_to_speech"]["enabled"] = False
+
+        if not isinstance(config["text_to_speech"]["enabled"], bool):
+            raise ValueError("enabled in text_to_speech must be of boolean type")
+
         # return retriever resource
         if 'retriever_resource' not in config or not config["retriever_resource"]:
             config["retriever_resource"] = {
@@ -317,6 +332,7 @@ class AppModelConfigService:
             "suggested_questions": config["suggested_questions"],
             "suggested_questions_after_answer": config["suggested_questions_after_answer"],
             "speech_to_text": config["speech_to_text"],
+            "text_to_speech": config["text_to_speech"],
             "retriever_resource": config["retriever_resource"],
             "more_like_this": config["more_like_this"],
             "sensitive_word_avoidance": config["sensitive_word_avoidance"],

api/services/audio_service.py

 import io
+from typing import Optional
 
 from core.model_manager import ModelManager
 from core.model_runtime.entities.model_entities import ModelType
-from services.errors.audio import (AudioTooLargeServiceError, NoAudioUploadedServiceError,
-                                   ProviderNotSupportSpeechToTextServiceError, UnsupportedAudioTypeServiceError)
+from services.errors.audio import (AudioTooLargeServiceError,
+                                   NoAudioUploadedServiceError,
+                                   ProviderNotSupportTextToSpeechServiceError,
+                                   ProviderNotSupportSpeechToTextServiceError,
+                                   UnsupportedAudioTypeServiceError)
 from werkzeug.datastructures import FileStorage
 
 FILE_SIZE = 15
 FILE_SIZE_LIMIT = FILE_SIZE * 1024 * 1024
-ALLOWED_EXTENSIONS = ['mp3', 'mp4', 'mpeg', 'mpga', 'm4a', 'wav', 'webm']
+ALLOWED_EXTENSIONS = ['mp3', 'mp4', 'mpeg', 'mpga', 'm4a', 'wav', 'webm', 'amr']
 
 
 class AudioService:
     @classmethod
-    def transcript(cls, tenant_id: str, file: FileStorage):
+    def transcript_asr(cls, tenant_id: str, file: FileStorage, end_user: Optional[str] = None):
         if file is None:
             raise NoAudioUploadedServiceError()
 
@@ -33,8 +37,26 @@ class AudioService:
             tenant_id=tenant_id,
             model_type=ModelType.SPEECH2TEXT
         )
+        if model_instance is None:
+            raise ProviderNotSupportSpeechToTextServiceError()
 
         buffer = io.BytesIO(file_content)
         buffer.name = 'temp.mp3'
 
-        return {"text": model_instance.invoke_speech2text(buffer)}
+        return {"text": model_instance.invoke_speech2text(file=buffer, user=end_user)}
+
+    @classmethod
+    def transcript_tts(cls, tenant_id: str, text: str, streaming: bool, end_user: Optional[str] = None):
+        model_manager = ModelManager()
+        model_instance = model_manager.get_default_model_instance(
+            tenant_id=tenant_id,
+            model_type=ModelType.TTS
+        )
+        if model_instance is None:
+            raise ProviderNotSupportTextToSpeechServiceError()
+
+        try:
+            audio_response = model_instance.invoke_tts(content_text=text.strip(), user=end_user, streaming=streaming)
+            return audio_response
+        except Exception as e:
+            raise e

api/services/errors/audio.py

@@ -9,5 +9,10 @@ class AudioTooLargeServiceError(Exception):
 class UnsupportedAudioTypeServiceError(Exception):
     pass
 
+
 class ProviderNotSupportSpeechToTextServiceError(Exception):
     pass
+
+
+class ProviderNotSupportTextToSpeechServiceError(Exception):
+    pass

web/app/components/app/chat/answer/index.tsx

@@ -13,6 +13,7 @@ import MoreInfo from '../more-info'
 import CopyBtn from '../copy-btn'
 import Thought from '../thought'
 import Citation from '../citation'
+import AudioBtn from '@/app/components/base/audio-btn'
 import { randomString } from '@/utils'
 import type { MessageRating } from '@/models/log'
 import Tooltip from '@/app/components/base/tooltip'
@@ -53,6 +54,7 @@ export type IAnswerProps = {
   dataSets?: DataSet[]
   isShowCitation?: boolean
   isShowCitationHitInfo?: boolean
+  isShowTextToSpeech?: boolean
   // Annotation props
   supportAnnotation?: boolean
   appId?: string
@@ -75,6 +77,7 @@ const Answer: FC<IAnswerProps> = ({
   citation,
   isShowCitation,
   isShowCitationHitInfo = false,
+  isShowTextToSpeech,
   supportAnnotation,
   appId,
   question,
@@ -322,7 +325,13 @@ const Answer: FC<IAnswerProps> = ({
                     className={cn(s.copyBtn, 'mr-1')}
                   />
                 )}
-                {(supportAnnotation && !item.isOpeningStatement) && (
+                {!item.isOpeningStatement && isShowTextToSpeech && (
+                  <AudioBtn
+                    value={content}
+                    className={cn(s.playBtn, 'mr-1')}
+                  />
+                )}
+                {(!item.isOpeningStatement && supportAnnotation) && (
                   <AnnotationCtrlBtn
                     appId={appId!}
                     messageId={id}

web/app/components/app/chat/index.tsx

@@ -57,6 +57,7 @@ export type IChatProps = {
   isShowSuggestion?: boolean
   suggestionList?: string[]
   isShowSpeechToText?: boolean
+  isShowTextToSpeech?: boolean
   isShowCitation?: boolean
   answerIcon?: ReactNode
   isShowConfigElem?: boolean
@@ -89,6 +90,7 @@ const Chat: FC<IChatProps> = ({
   isShowSuggestion,
   suggestionList,
   isShowSpeechToText,
+  isShowTextToSpeech,
   isShowCitation,
   answerIcon,
   isShowConfigElem,
@@ -222,6 +224,7 @@ const Chat: FC<IChatProps> = ({
               dataSets={dataSets}
               isShowCitation={isShowCitation}
               isShowCitationHitInfo={isShowCitationHitInfo}
+              isShowTextToSpeech={isShowTextToSpeech}
               supportAnnotation={supportAnnotation}
               appId={appId}
               question={chatList[index - 1]?.content}

web/app/components/app/chat/style.module.css

@@ -39,6 +39,7 @@
 }
 
 .copyBtn,
+.playBtn,
 .annotationBtn {
   display: none;
 }
@@ -65,6 +66,7 @@
 }
 
 .answerWrap:hover .copyBtn,
+.answerWrap:hover .playBtn,
 .answerWrap:hover .annotationBtn {
   display: block;
 }

web/app/components/app/configuration/config/feature/choose-feature/feature-item/style.module.css

@@ -29,6 +29,11 @@
   background-image: url(./preview-imgs/speech-to-text.svg);
 }
 
+.textToSpeechPreview {
+  @apply shadow-lg rounded-lg;
+  background-image: url(./preview-imgs/text-to-audio-preview-assistant@2x.png);
+}
+
 .citationPreview {
   background-image: url(./preview-imgs/citation.svg);
 }

web/app/components/app/configuration/config/feature/choose-feature/index.tsx

@@ -7,7 +7,7 @@ import MoreLikeThisIcon from '../../../base/icons/more-like-this-icon'
 import FeatureItem from './feature-item'
 import Modal from '@/app/components/base/modal'
 import SuggestedQuestionsAfterAnswerIcon from '@/app/components/app/configuration/base/icons/suggested-questions-after-answer-icon'
-import { Microphone01 } from '@/app/components/base/icons/src/vender/solid/mediaAndDevices'
+import { Microphone01, Speaker } from '@/app/components/base/icons/src/vender/solid/mediaAndDevices'
 import { Citations } from '@/app/components/base/icons/src/vender/solid/editor'
 import { FileSearch02 } from '@/app/components/base/icons/src/vender/solid/files'
 import { MessageFast } from '@/app/components/base/icons/src/vender/solid/communication'
@@ -16,6 +16,7 @@ type IConfig = {
   moreLikeThis: boolean
   suggestedQuestionsAfterAnswer: boolean
   speechToText: boolean
+  textToSpeech: boolean
   citation: boolean
   moderation: boolean
   annotation: boolean
@@ -27,6 +28,7 @@ export type IChooseFeatureProps = {
   config: IConfig
   isChatApp: boolean
   onChange: (key: string, value: boolean) => void
+  showTextToSpeechItem?: boolean
   showSpeechToTextItem?: boolean
 }
 
@@ -42,6 +44,7 @@ const ChooseFeature: FC<IChooseFeatureProps> = ({
   isChatApp,
   config,
   onChange,
+  showTextToSpeechItem,
   showSpeechToTextItem,
 }) => {
   const { t } = useTranslation()
@@ -78,6 +81,18 @@ const ChooseFeature: FC<IChooseFeatureProps> = ({
                 value={config.suggestedQuestionsAfterAnswer}
                 onChange={value => onChange('suggestedQuestionsAfterAnswer', value)}
               />
+              {
+                showTextToSpeechItem && (
+                  <FeatureItem
+                    icon={<Speaker className='w-4 h-4 text-[#7839EE]' />}
+                    previewImgClassName='textToSpeechPreview'
+                    title={t('appDebug.feature.textToSpeech.title')}
+                    description={t('appDebug.feature.textToSpeech.description')}
+                    value={config.textToSpeech}
+                    onChange={value => onChange('textToSpeech', value)}
+                  />
+                )
+              }
               {
                 showSpeechToTextItem && (
                   <FeatureItem
@@ -114,6 +129,18 @@ const ChooseFeature: FC<IChooseFeatureProps> = ({
                 value={config.moreLikeThis}
                 onChange={value => onChange('moreLikeThis', value)}
               />
+              {
+                showTextToSpeechItem && (
+                  <FeatureItem
+                    icon={<Speaker className='w-4 h-4 text-[#7839EE]' />}
+                    previewImgClassName='textToSpeechPreview'
+                    title={t('appDebug.feature.textToSpeech.title')}
+                    description={t('appDebug.feature.textToSpeech.description')}
+                    value={config.textToSpeech}
+                    onChange={value => onChange('textToSpeech', value)}
+                  />
+                )
+              }
             </>
           </FeatureGroup>
         )}

web/app/components/app/configuration/config/feature/use-feature.tsx

@@ -9,6 +9,8 @@ function useFeature({
   setSuggestedQuestionsAfterAnswer,
   speechToText,
   setSpeechToText,
+  textToSpeech,
+  setTextToSpeech,
   citation,
   setCitation,
   annotation,
@@ -24,6 +26,8 @@ function useFeature({
   setSuggestedQuestionsAfterAnswer: (suggestedQuestionsAfterAnswer: boolean) => void
   speechToText: boolean
   setSpeechToText: (speechToText: boolean) => void
+  textToSpeech: boolean
+  setTextToSpeech: (textToSpeech: boolean) => void
   citation: boolean
   setCitation: (citation: boolean) => void
   annotation: boolean
@@ -48,6 +52,7 @@ function useFeature({
     moreLikeThis,
     suggestedQuestionsAfterAnswer,
     speechToText,
+    textToSpeech,
     citation,
     annotation,
     moderation,
@@ -69,6 +74,9 @@ function useFeature({
       case 'speechToText':
         setSpeechToText(value)
         break
+      case 'textToSpeech':
+        setTextToSpeech(value)
+        break
       case 'citation':
         setCitation(value)
         break

web/app/components/app/configuration/config/index.tsx

@@ -19,7 +19,7 @@ import AdvancedModeWaring from '@/app/components/app/configuration/prompt-mode/a
 import ConfigContext from '@/context/debug-configuration'
 import ConfigPrompt from '@/app/components/app/configuration/config-prompt'
 import ConfigVar from '@/app/components/app/configuration/config-var'
-import { type CitationConfig, type ModelConfig, type ModerationConfig, type MoreLikeThisConfig, PromptMode, type PromptVariable, type SpeechToTextConfig, type SuggestedQuestionsAfterAnswerConfig } from '@/models/debug'
+import { type CitationConfig, type ModelConfig, type ModerationConfig, type MoreLikeThisConfig, PromptMode, type PromptVariable, type SpeechToTextConfig, type SuggestedQuestionsAfterAnswerConfig, type TextToSpeechConfig } from '@/models/debug'
 import { AppType, ModelModeType } from '@/types/app'
 import { useModalContext } from '@/context/modal-context'
 import ConfigParamModal from '@/app/components/app/configuration/toolbox/annotation/config-param-modal'
@@ -51,6 +51,8 @@ const Config: FC = () => {
     setSuggestedQuestionsAfterAnswerConfig,
     speechToTextConfig,
     setSpeechToTextConfig,
+    textToSpeechConfig,
+    setTextToSpeechConfig,
     citationConfig,
     setCitationConfig,
     annotationConfig,
@@ -60,6 +62,7 @@ const Config: FC = () => {
   } = useContext(ConfigContext)
   const isChatApp = mode === AppType.chat
   const { data: speech2textDefaultModel } = useDefaultModel(4)
+  const { data: text2speechDefaultModel } = useDefaultModel(5)
   const { setShowModerationSettingModal } = useModalContext()
 
   const promptTemplate = modelConfig.configs.prompt_template
@@ -111,6 +114,12 @@ const Config: FC = () => {
         draft.enabled = value
       }))
     },
+    textToSpeech: textToSpeechConfig.enabled,
+    setTextToSpeech: (value) => {
+      setTextToSpeechConfig(produce(textToSpeechConfig, (draft: TextToSpeechConfig) => {
+        draft.enabled = value
+      }))
+    },
     citation: citationConfig.enabled,
     setCitation: (value) => {
       setCitationConfig(produce(citationConfig, (draft: CitationConfig) => {
@@ -173,7 +182,7 @@ const Config: FC = () => {
     setAnnotationConfig,
   })
 
-  const hasChatConfig = isChatApp && (featureConfig.openingStatement || featureConfig.suggestedQuestionsAfterAnswer || (featureConfig.speechToText && !!speech2textDefaultModel) || featureConfig.citation)
+  const hasChatConfig = isChatApp && (featureConfig.openingStatement || featureConfig.suggestedQuestionsAfterAnswer || (featureConfig.speechToText && !!speech2textDefaultModel) || (featureConfig.textToSpeech && !!text2speechDefaultModel) || featureConfig.citation)
   const hasToolbox = moderationConfig.enabled || featureConfig.annotation
 
   const wrapRef = useRef<HTMLDivElement>(null)
@@ -207,6 +216,7 @@ const Config: FC = () => {
             config={featureConfig}
             onChange={handleFeatureChange}
             showSpeechToTextItem={!!speech2textDefaultModel}
+            showTextToSpeechItem={!!text2speechDefaultModel}
           />
         )}
 
@@ -255,16 +265,21 @@ const Config: FC = () => {
                 }
               }
               isShowSuggestedQuestionsAfterAnswer={featureConfig.suggestedQuestionsAfterAnswer}
+              isShowTextToSpeech={featureConfig.textToSpeech && !!text2speechDefaultModel}
               isShowSpeechText={featureConfig.speechToText && !!speech2textDefaultModel}
               isShowCitation={featureConfig.citation}
             />
           )
         }
 
-        {/* TextnGeneration config */}
-        {moreLikeThisConfig.enabled && (
-          <ExperienceEnchanceGroup />
-        )}
+        {/* TextnGeneration config */}{
+          !hasChatConfig && (
+            <ExperienceEnchanceGroup
+              isShowMoreLike={moreLikeThisConfig.enabled}
+              isShowTextToSpeech={featureConfig.textToSpeech && !!text2speechDefaultModel}
+            />
+          )
+        }
 
         {/* Toolbox */}
         {

web/app/components/app/configuration/debug/index.tsx

@@ -56,6 +56,7 @@ const Debug: FC<IDebug> = ({
     suggestedQuestions,
     suggestedQuestionsAfterAnswerConfig,
     speechToTextConfig,
+    textToSpeechConfig,
     citationConfig,
     moderationConfig,
     moreLikeThisConfig,
@@ -73,6 +74,7 @@ const Debug: FC<IDebug> = ({
     annotationConfig,
   } = useContext(ConfigContext)
   const { data: speech2textDefaultModel } = useDefaultModel(4)
+  const { data: text2speechDefaultModel } = useDefaultModel(5)
   const [chatList, setChatList, getChatList] = useGetState<IChatItem[]>([])
   const chatListDomRef = useRef<HTMLDivElement>(null)
   const { data: fileUploadConfigResponse } = useSWR({ url: '/files/upload' }, fetchFileUploadConfig)
@@ -233,6 +235,9 @@ const Debug: FC<IDebug> = ({
       setChatList(newListWithAnswer)
     }
     const postModelConfig: BackendModelConfig = {
+      text_to_speech: {
+        enabled: false,
+      },
       pre_prompt: !isAdvancedMode ? modelConfig.configs.prompt_template : '',
       prompt_type: promptMode,
       chat_prompt_config: {},
@@ -514,6 +519,9 @@ const Debug: FC<IDebug> = ({
     const contextVar = modelConfig.configs.prompt_variables.find(item => item.is_context_var)?.key
 
     const postModelConfig: BackendModelConfig = {
+      text_to_speech: {
+        enabled: false,
+      },
       pre_prompt: !isAdvancedMode ? modelConfig.configs.prompt_template : '',
       prompt_type: promptMode,
       chat_prompt_config: {},
@@ -657,6 +665,7 @@ const Debug: FC<IDebug> = ({
                   isShowSuggestion={doShowSuggestion}
                   suggestionList={suggestQuestions}
                   isShowSpeechToText={speechToTextConfig.enabled && !!speech2textDefaultModel}
+                  isShowTextToSpeech={textToSpeechConfig.enabled && !!text2speechDefaultModel}
                   isShowCitation={citationConfig.enabled}
                   isShowCitationHitInfo
                   isShowPromptLog
@@ -682,6 +691,7 @@ const Debug: FC<IDebug> = ({
                 className="mt-2"
                 content={completionRes}
                 isLoading={!completionRes && isResponsing}
+                isShowTextToSpeech={textToSpeechConfig.enabled && !!text2speechDefaultModel}
                 isResponsing={isResponsing}
                 isInstalledApp={false}
                 messageId={messageId}

web/app/components/app/configuration/features/chat-group/index.tsx

@@ -7,6 +7,7 @@ import type { IOpeningStatementProps } from './opening-statement'
 import OpeningStatement from './opening-statement'
 import SuggestedQuestionsAfterAnswer from './suggested-questions-after-answer'
 import SpeechToText from './speech-to-text'
+import TextToSpeech from './text-to-speech'
 import Citation from './citation'
 /*
 * Include
@@ -19,6 +20,7 @@ type ChatGroupProps = {
   openingStatementConfig: IOpeningStatementProps
   isShowSuggestedQuestionsAfterAnswer: boolean
   isShowSpeechText: boolean
+  isShowTextToSpeech: boolean
   isShowCitation: boolean
 }
 const ChatGroup: FC<ChatGroupProps> = ({
@@ -26,6 +28,7 @@ const ChatGroup: FC<ChatGroupProps> = ({
   openingStatementConfig,
   isShowSuggestedQuestionsAfterAnswer,
   isShowSpeechText,
+  isShowTextToSpeech,
   isShowCitation,
 }) => {
   const { t } = useTranslation()
@@ -40,6 +43,11 @@ const ChatGroup: FC<ChatGroupProps> = ({
         {isShowSuggestedQuestionsAfterAnswer && (
           <SuggestedQuestionsAfterAnswer />
         )}
+        {
+          isShowTextToSpeech && (
+            <TextToSpeech />
+          )
+        }
         {
           isShowSpeechText && (
             <SpeechToText />

web/app/components/app/configuration/features/chat-group/speech-to-text/index.tsx

@@ -4,7 +4,7 @@ import { useTranslation } from 'react-i18next'
 import Panel from '@/app/components/app/configuration/base/feature-panel'
 import { Microphone01 } from '@/app/components/base/icons/src/vender/solid/mediaAndDevices'
 
-const SuggestedQuestionsAfterAnswer: FC = () => {
+const SpeechToTextConfig: FC = () => {
   const { t } = useTranslation()
 
   return (
@@ -22,4 +22,4 @@ const SuggestedQuestionsAfterAnswer: FC = () => {
     />
   )
 }
-export default React.memo(SuggestedQuestionsAfterAnswer)
+export default React.memo(SpeechToTextConfig)

web/app/components/app/configuration/features/chat-group/text-to-speech/index.tsx

+'use client'
+import React, { type FC } from 'react'
+import { useTranslation } from 'react-i18next'
+import Panel from '@/app/components/app/configuration/base/feature-panel'
+import { Speaker } from '@/app/components/base/icons/src/vender/solid/mediaAndDevices'
+
+const TextToSpeech: FC = () => {
+  const { t } = useTranslation()
+
+  return (
+    <Panel
+      title={
+        <div className='flex items-center gap-2'>
+          <div>{t('appDebug.feature.textToSpeech.title')}</div>
+        </div>
+      }
+      headerIcon={<Speaker className='w-4 h-4 text-[#7839EE]' />}
+      headerRight={
+        <div className='text-xs text-gray-500'>{t('appDebug.feature.textToSpeech.resDes')}</div>
+      }
+      noBodySpacing
+    />
+  )
+}
+export default React.memo(TextToSpeech)

web/app/components/app/configuration/features/experience-enchance-group/index.tsx

@@ -3,19 +3,40 @@ import type { FC } from 'react'
 import React from 'react'
 import { useTranslation } from 'react-i18next'
 import GroupName from '../../base/group-name'
+import TextToSpeech from '../chat-group/text-to-speech'
 import MoreLikeThis from './more-like-this'
 
 /*
 * Include
 * 1. More like this
 */
-const ExperienceEnchanceGroup: FC = () => {
+
+type ExperienceGroupProps = {
+  isShowTextToSpeech: boolean
+  isShowMoreLike: boolean
+}
+
+const ExperienceEnchanceGroup: FC<ExperienceGroupProps> = ({
+  isShowTextToSpeech,
+  isShowMoreLike,
+}) => {
   const { t } = useTranslation()
 
   return (
     <div className='mt-7'>
-      <GroupName name={t('appDebug.feature.groupExperience.title')} />
-      <MoreLikeThis />
+      <GroupName name={t('appDebug.feature.groupExperience.title')}/>
+      <div className='space-y-3'>
+        {
+          isShowMoreLike && (
+            <MoreLikeThis/>
+          )
+        }
+        {
+          isShowTextToSpeech && (
+            <TextToSpeech/>
+          )
+        }
+      </div>
     </div>
   )
 }

web/app/components/app/configuration/index.tsx

@@ -91,6 +91,9 @@ const Configuration: FC = () => {
   const [speechToTextConfig, setSpeechToTextConfig] = useState<MoreLikeThisConfig>({
     enabled: false,
   })
+  const [textToSpeechConfig, setTextToSpeechConfig] = useState<MoreLikeThisConfig>({
+    enabled: false,
+  })
   const [citationConfig, setCitationConfig] = useState<MoreLikeThisConfig>({
     enabled: false,
   })
@@ -140,6 +143,7 @@ const Configuration: FC = () => {
     more_like_this: null,
     suggested_questions_after_answer: null,
     speech_to_text: null,
+    text_to_speech: null,
     retriever_resource: null,
     sensitive_word_avoidance: null,
     dataSets: [],
@@ -232,6 +236,9 @@ const Configuration: FC = () => {
     setSpeechToTextConfig(modelConfig.speech_to_text || {
       enabled: false,
     })
+    setTextToSpeechConfig(modelConfig.text_to_speech || {
+      enabled: false,
+    })
     setCitationConfig(modelConfig.retriever_resource || {
       enabled: false,
     })
@@ -396,6 +403,9 @@ const Configuration: FC = () => {
         if (modelConfig.speech_to_text)
           setSpeechToTextConfig(modelConfig.speech_to_text)
 
+        if (modelConfig.text_to_speech)
+          setTextToSpeechConfig(modelConfig.text_to_speech)
+
         if (modelConfig.retriever_resource)
           setCitationConfig(modelConfig.retriever_resource)
 
@@ -444,6 +454,7 @@ const Configuration: FC = () => {
             more_like_this: modelConfig.more_like_this,
             suggested_questions_after_answer: modelConfig.suggested_questions_after_answer,
             speech_to_text: modelConfig.speech_to_text,
+            text_to_speech: modelConfig.text_to_speech,
             retriever_resource: modelConfig.retriever_resource,
             sensitive_word_avoidance: modelConfig.sensitive_word_avoidance,
             external_data_tools: modelConfig.external_data_tools,
@@ -559,6 +570,7 @@ const Configuration: FC = () => {
       more_like_this: moreLikeThisConfig,
       suggested_questions_after_answer: suggestedQuestionsAfterAnswerConfig,
       speech_to_text: speechToTextConfig,
+      text_to_speech: textToSpeechConfig,
       retriever_resource: citationConfig,
       sensitive_word_avoidance: moderationConfig,
       agent_mode: {
@@ -593,6 +605,7 @@ const Configuration: FC = () => {
       draft.more_like_this = moreLikeThisConfig
       draft.suggested_questions_after_answer = suggestedQuestionsAfterAnswerConfig
       draft.speech_to_text = speechToTextConfig
+      draft.text_to_speech = textToSpeechConfig
       draft.retriever_resource = citationConfig
       draft.dataSets = dataSets
     })
@@ -662,6 +675,8 @@ const Configuration: FC = () => {
       setSuggestedQuestionsAfterAnswerConfig,
       speechToTextConfig,
       setSpeechToTextConfig,
+      textToSpeechConfig,
+      setTextToSpeechConfig,
       citationConfig,
       setCitationConfig,
       annotationConfig,

web/app/components/app/log/list.tsx

@@ -297,6 +297,7 @@ function DetailPanel<T extends ChatConversationFullDetailResponse | CompletionCo
           feedback={detail.message.feedbacks.find((item: any) => item.from_source === 'admin')}
           onFeedback={feedback => onFeedback(detail.message.id, feedback)}
           supportAnnotation
+          isShowTextToSpeech
           appId={appDetail?.id}
           varList={varList}
         />
@@ -310,6 +311,7 @@ function DetailPanel<T extends ChatConversationFullDetailResponse | CompletionCo
             displayScene='console'
             isShowPromptLog
             supportAnnotation
+            isShowTextToSpeech
             appId={appDetail?.id}
             onChatListChange={setItems}
           />

web/app/components/app/text-generate/item/index.tsx

@@ -12,6 +12,7 @@ import PromptLog from '@/app/components/app/chat/log'
 import { Markdown } from '@/app/components/base/markdown'
 import Loading from '@/app/components/base/loading'
 import Toast from '@/app/components/base/toast'
+import AudioBtn from '@/app/components/base/audio-btn'
 import type { Feedbacktype } from '@/app/components/app/chat/type'
 import { fetchMoreLikeThis, updateFeedback } from '@/service/share'
 import { Clipboard, File02 } from '@/app/components/base/icons/src/vender/line/files'
@@ -45,6 +46,7 @@ export type IGenerationItemProps = {
   controlClearMoreLikeThis?: number
   supportFeedback?: boolean
   supportAnnotation?: boolean
+  isShowTextToSpeech?: boolean
   appId?: string
   varList?: { label: string; value: string | number | object }[]
 }
@@ -90,6 +92,7 @@ const GenerationItem: FC<IGenerationItemProps> = ({
   controlClearMoreLikeThis,
   supportFeedback,
   supportAnnotation,
+  isShowTextToSpeech,
   appId,
   varList,
 }) => {
@@ -124,6 +127,7 @@ const GenerationItem: FC<IGenerationItemProps> = ({
     isLoading: isQuerying,
     feedback: childFeedback,
     onSave,
+    isShowTextToSpeech,
     isMobile,
     isInstalledApp,
     installedAppId,
@@ -366,8 +370,17 @@ const GenerationItem: FC<IGenerationItemProps> = ({
                   <div className='ml-1'>
                     {ratingContent}
                   </div>
-                )
-                }
+                )}
+
+                {isShowTextToSpeech && (
+                  <>
+                    <div className='ml-2 mr-2 h-[14px] w-[1px] bg-gray-200'></div>
+                    <AudioBtn
+                      value={content}
+                      className={'mr-1'}
+                    />
+                  </>
+                )}
               </div>
               <div className='text-xs text-gray-500'>{content?.length} {t('common.unit.char')}</div>
             </div>

web/app/components/app/text-generate/saved-items/index.tsx

@@ -9,9 +9,11 @@ import type { SavedMessage } from '@/models/debug'
 import { Markdown } from '@/app/components/base/markdown'
 import { SimpleBtn, copyIcon } from '@/app/components/app/text-generate/item'
 import Toast from '@/app/components/base/toast'
+import AudioBtn from '@/app/components/base/audio-btn'
 
 export type ISavedItemsProps = {
   className?: string
+  isShowTextToSpeech?: boolean
   list: SavedMessage[]
   onRemove: (id: string) => void
   onStartCreateContent: () => void
@@ -25,6 +27,7 @@ const removeIcon = (
 
 const SavedItems: FC<ISavedItemsProps> = ({
   className,
+  isShowTextToSpeech,
   list,
   onRemove,
   onStartCreateContent,
@@ -69,6 +72,16 @@ const SavedItems: FC<ISavedItemsProps> = ({
                     {removeIcon}
                     <div>{t('common.operation.remove')}</div>
                   </SimpleBtn>
+
+                  {isShowTextToSpeech && (
+                    <>
+                      <div className='ml-2 mr-2 h-[14px] w-[1px] bg-gray-200'></div>
+                      <AudioBtn
+                        value={answer}
+                        className={'mr-1'}
+                      />
+                    </>
+                  )}
                 </div>
                 <div className='text-xs text-gray-500'>{answer?.length} {t('common.unit.char')}</div>
               </div>

web/app/components/base/audio-btn/index.tsx

+'use client'
+import { useRef, useState } from 'react'
+import { t } from 'i18next'
+import { useParams, usePathname } from 'next/navigation'
+import s from './style.module.css'
+import Tooltip from '@/app/components/base/tooltip'
+import { randomString } from '@/utils'
+import { textToAudio } from '@/service/share'
+
+type AudioBtnProps = {
+  value: string
+  className?: string
+}
+
+const AudioBtn = ({
+  value,
+  className,
+}: AudioBtnProps) => {
+  const audioRef = useRef<HTMLAudioElement | null>(null)
+  const [isPlaying, setIsPlaying] = useState(false)
+  const [isPause, setPause] = useState(false)
+  const [hasEnded, setHasEnded] = useState(false)
+  const selector = useRef(`play-tooltip-${randomString(4)}`)
+  const params = useParams()
+  const pathname = usePathname()
+  const removeCodeBlocks = (inputText: any) => {
+    const codeBlockRegex = /```[\s\S]*?```/g
+    return inputText.replace(codeBlockRegex, '')
+  }
+
+  const playAudio = async () => {
+    const formData = new FormData()
+    if (value !== '') {
+      formData.append('text', removeCodeBlocks(value))
+
+      let url = '/universal-chat/text-to-audio'
+      let isPublic = false
+
+      if (params.token) {
+        url = '/text-to-audio'
+        isPublic = true
+      }
+      else if (params.appId) {
+        if (pathname.search('explore/installed') > -1)
+          url = `/installed-apps/${params.appId}/text-to-audio`
+        else
+          url = `/apps/${params.appId}/text-to-audio`
+      }
+
+      try {
+        const audioResponse = await textToAudio(url, isPublic, formData)
+        const blob_bytes = Buffer.from(audioResponse.data, 'latin1')
+        const blob = new Blob([blob_bytes], { type: 'audio/wav' })
+        const audioUrl = URL.createObjectURL(blob)
+        const audio = new Audio(audioUrl)
+        audioRef.current = audio
+        audio.play().then(() => {
+          setIsPlaying(true)
+        }).catch(() => {
+          setIsPlaying(false)
+          URL.revokeObjectURL(audioUrl)
+        })
+        audio.onended = () => setHasEnded(true)
+      }
+      catch (error) {
+        setIsPlaying(false)
+        console.error('Error playing audio:', error)
+      }
+    }
+  }
+
+  const togglePlayPause = () => {
+    if (audioRef.current) {
+      if (isPlaying) {
+        setPause(true)
+        audioRef.current.pause()
+      }
+      else if (!hasEnded) {
+        setPause(false)
+        audioRef.current.play()
+      }
+      else if (!isPlaying) {
+        playAudio().then()
+      }
+      setIsPlaying(prevIsPlaying => !prevIsPlaying)
+    }
+    else {
+      playAudio().then()
+    }
+  }
+
+  return (
+    <div className={`${(isPlaying && !hasEnded) ? 'mr-1' : className}`}>
+      <Tooltip
+        selector={selector.current}
+        content={(!isPause ? ((isPlaying && !hasEnded) ? t('appApi.playing') : t('appApi.play')) : t('appApi.pause')) as string}
+        className='z-10'
+      >
+        <div
+          className={'box-border p-0.5 flex items-center justify-center rounded-md bg-white cursor-pointer'}
+          style={{ boxShadow: '0px 4px 8px -2px rgba(16, 24, 40, 0.1), 0px 2px 4px -2px rgba(16, 24, 40, 0.06)' }}
+          onClick={togglePlayPause}>
+          <div className={`w-6 h-6 rounded-md hover:bg-gray-50 ${!isPause ? ((isPlaying && !hasEnded) ? s.playIcon : s.stopIcon) : s.pauseIcon}`}></div>
+        </div>
+      </Tooltip>
+    </div>
+  )
+}
+
+export default AudioBtn

web/app/components/base/audio-btn/style.module.css

+.playIcon {
+  background-image: url(~@/app/components/develop/secret-key/assets/play.svg);
+  background-position: center;
+  background-repeat: no-repeat;
+}
+.pauseIcon {
+  background-image: url(~@/app/components/develop/secret-key/assets/pause.svg);
+  background-position: center;
+  background-repeat: no-repeat;
+}
+
+.stopIcon {
+  background-position: center;
+  background-repeat: no-repeat;
+  background-image: url(~@/app/components/develop/secret-key/assets/stop.svg);
+}
\ No newline at end of file

web/app/components/base/icons/assets/vender/line/mediaAndDevices/speaker.svg

+<svg width="16" height="16" viewBox="0 0 16 16" fill="none" xmlns="http://www.w3.org/2000/svg">
+<g clip-path="url(#clip0_109_6694)">
+<path fill-rule="evenodd" clip-rule="evenodd" d="M0 2.86666C0 2.05664 0.656649 1.39999 1.46667 1.39999H5.86667C6.67668 1.39999 7.33333 2.05664 7.33333 2.86666C7.33333 3.27167 7.00501 3.59999 6.6 3.59999C6.19499 3.59999 5.86667 3.27167 5.86667 2.86666H4.4V7.99999C4.80501 7.99999 5.13333 8.32831 5.13333 8.73332C5.13333 9.13833 4.80501 9.46666 4.4 9.46666H2.93333C2.52832 9.46666 2.2 9.13833 2.2 8.73332C2.2 8.32831 2.52832 7.99999 2.93333 7.99999V2.86666H1.46667C1.46667 3.27167 1.13834 3.59999 0.733333 3.59999C0.328324 3.59999 0 3.27167 0 2.86666Z" fill="#444CE7"/>
+<path d="M13.8205 0.782296C13.7434 0.62811 13.5233 0.62811 13.4462 0.782296C12.9664 1.74206 12.8754 1.83302 11.9156 2.3129C11.7615 2.39 11.7615 2.61003 11.9156 2.68712C12.8754 3.167 12.9664 3.25797 13.4462 4.21773C13.5233 4.37191 13.7434 4.37191 13.8205 4.21773C14.3003 3.25797 14.3913 3.167 15.3511 2.68712C15.5053 2.61003 15.5053 2.39 15.3511 2.3129C14.3913 1.83302 14.3003 1.74206 13.8205 0.782296Z" fill="#444CE7"/>
+<path d="M9.79394 2.25319C9.71404 2.09337 9.48596 2.09337 9.40605 2.25319C9.04994 2.96543 8.96544 3.04993 8.2532 3.40605C8.09338 3.48595 8.09338 3.71402 8.2532 3.79393C8.96544 4.15005 9.04994 4.23455 9.40606 4.94679C9.48596 5.10661 9.71404 5.10661 9.79394 4.94679C10.1501 4.23455 10.2346 4.15005 10.9468 3.79393C11.1066 3.71402 11.1066 3.48595 10.9468 3.40605C10.2346 3.04993 10.1501 2.96543 9.79394 2.25319Z" fill="#444CE7"/>
+<path d="M2.75377 11.049C2.67668 10.8948 2.45665 10.8948 2.37956 11.049C1.89969 12.0087 1.80872 12.0997 0.848971 12.5796C0.694788 12.6566 0.694787 12.8767 0.848971 12.9538C1.80872 13.4336 1.89969 13.5246 2.37956 14.4844C2.45665 14.6385 2.67668 14.6385 2.75377 14.4844C3.23365 13.5246 3.32461 13.4336 4.28436 12.9538C4.43855 12.8767 4.43855 12.6566 4.28436 12.5796C3.32461 12.0997 3.23365 12.0087 2.75377 11.049Z" fill="#444CE7"/>
+<path fill-rule="evenodd" clip-rule="evenodd" d="M14.6741 8.65106C14.8886 8.50146 15.1837 8.55405 15.3333 8.76853C15.7614 9.38226 16.0125 10.1292 16.0125 10.9333C16.0125 11.7375 15.7614 12.4844 15.3333 13.0981C15.1837 13.3126 14.8886 13.3652 14.6741 13.2156C14.4596 13.066 14.407 12.7708 14.5567 12.5564C14.8775 12.0964 15.0656 11.5375 15.0656 10.9333C15.0656 10.3291 14.8775 9.77025 14.5567 9.31028C14.407 9.09581 14.4596 8.80066 14.6741 8.65106Z" fill="#444CE7"/>
+<path fill-rule="evenodd" clip-rule="evenodd" d="M12.5674 6.53771C12.794 6.51987 13.0155 6.61161 13.1632 6.78449C13.2954 6.93929 13.3164 7.12549 13.3244 7.21587C13.3334 7.31718 13.3334 7.44301 13.3333 7.57103C13.3333 7.57691 13.3333 7.58278 13.3333 7.58866L13.3333 14.3C13.3334 14.428 13.3334 14.5539 13.3244 14.6552C13.3164 14.7455 13.2954 14.9317 13.1632 15.0865C13.0155 15.2594 12.794 15.3512 12.5674 15.3333C12.3644 15.3173 12.2179 15.2005 12.1484 15.1423C12.0704 15.077 11.9814 14.988 11.8909 14.8975L10.3795 13.3861C10.3357 13.3423 10.3137 13.3205 10.2971 13.3053L10.2958 13.3041L10.2941 13.3041C10.2716 13.303 10.2407 13.3029 10.1787 13.3029L9.34101 13.3029C9.22151 13.3029 9.10513 13.3029 9.00657 13.2949C8.89833 13.286 8.77062 13.2652 8.6421 13.1997C8.46392 13.1089 8.31906 12.964 8.22827 12.7859C8.16279 12.6574 8.14192 12.5296 8.13308 12.4214C8.12503 12.3228 8.12504 12.2065 8.12505 12.087V9.79916C8.12505 9.79413 8.12505 9.78909 8.12505 9.78406C8.12504 9.66456 8.12503 9.54819 8.13308 9.44963C8.14192 9.34139 8.16279 9.21368 8.22827 9.08517C8.31906 8.90699 8.46392 8.76212 8.6421 8.67133C8.77062 8.60585 8.89833 8.58498 9.00657 8.57614C9.10512 8.56809 9.2215 8.5681 9.341 8.56812C9.34603 8.56812 9.35106 8.56812 9.3561 8.56812H10.1787C10.2407 8.56812 10.2716 8.56801 10.2941 8.56698L10.2958 8.5669L10.2971 8.56575C10.3137 8.55058 10.3357 8.52877 10.3795 8.48491L11.8784 6.98602C11.8826 6.98186 11.8867 6.97771 11.8909 6.97355C11.9814 6.88302 12.0704 6.79403 12.1484 6.72874C12.2179 6.67049 12.3644 6.55368 12.5674 6.53771Z" fill="#444CE7"/>
+</g>
+<defs>
+<clipPath id="clip0_109_6694">
+<rect width="16" height="16" fill="white"/>
+</clipPath>
+</defs>
+</svg>

web/app/components/base/icons/assets/vender/solid/mediaAndDevices/speaker.svg

+<svg width="16" height="16" viewBox="0 0 16 16" fill="none" xmlns="http://www.w3.org/2000/svg">
+<g clip-path="url(#clip0_109_6694)">
+<path fill-rule="evenodd" clip-rule="evenodd" d="M0 2.86666C0 2.05664 0.656649 1.39999 1.46667 1.39999H5.86667C6.67668 1.39999 7.33333 2.05664 7.33333 2.86666C7.33333 3.27167 7.00501 3.59999 6.6 3.59999C6.19499 3.59999 5.86667 3.27167 5.86667 2.86666H4.4V7.99999C4.80501 7.99999 5.13333 8.32831 5.13333 8.73332C5.13333 9.13833 4.80501 9.46666 4.4 9.46666H2.93333C2.52832 9.46666 2.2 9.13833 2.2 8.73332C2.2 8.32831 2.52832 7.99999 2.93333 7.99999V2.86666H1.46667C1.46667 3.27167 1.13834 3.59999 0.733333 3.59999C0.328324 3.59999 0 3.27167 0 2.86666Z" fill="#444CE7"/>
+<path d="M13.8205 0.782296C13.7434 0.62811 13.5233 0.62811 13.4462 0.782296C12.9664 1.74206 12.8754 1.83302 11.9156 2.3129C11.7615 2.39 11.7615 2.61003 11.9156 2.68712C12.8754 3.167 12.9664 3.25797 13.4462 4.21773C13.5233 4.37191 13.7434 4.37191 13.8205 4.21773C14.3003 3.25797 14.3913 3.167 15.3511 2.68712C15.5053 2.61003 15.5053 2.39 15.3511 2.3129C14.3913 1.83302 14.3003 1.74206 13.8205 0.782296Z" fill="#444CE7"/>
+<path d="M9.79394 2.25319C9.71404 2.09337 9.48596 2.09337 9.40605 2.25319C9.04994 2.96543 8.96544 3.04993 8.2532 3.40605C8.09338 3.48595 8.09338 3.71402 8.2532 3.79393C8.96544 4.15005 9.04994 4.23455 9.40606 4.94679C9.48596 5.10661 9.71404 5.10661 9.79394 4.94679C10.1501 4.23455 10.2346 4.15005 10.9468 3.79393C11.1066 3.71402 11.1066 3.48595 10.9468 3.40605C10.2346 3.04993 10.1501 2.96543 9.79394 2.25319Z" fill="#444CE7"/>
+<path d="M2.75377 11.049C2.67668 10.8948 2.45665 10.8948 2.37956 11.049C1.89969 12.0087 1.80872 12.0997 0.848971 12.5796C0.694788 12.6566 0.694787 12.8767 0.848971 12.9538C1.80872 13.4336 1.89969 13.5246 2.37956 14.4844C2.45665 14.6385 2.67668 14.6385 2.75377 14.4844C3.23365 13.5246 3.32461 13.4336 4.28436 12.9538C4.43855 12.8767 4.43855 12.6566 4.28436 12.5796C3.32461 12.0997 3.23365 12.0087 2.75377 11.049Z" fill="#444CE7"/>
+<path fill-rule="evenodd" clip-rule="evenodd" d="M14.6741 8.65106C14.8886 8.50146 15.1837 8.55405 15.3333 8.76853C15.7614 9.38226 16.0125 10.1292 16.0125 10.9333C16.0125 11.7375 15.7614 12.4844 15.3333 13.0981C15.1837 13.3126 14.8886 13.3652 14.6741 13.2156C14.4596 13.066 14.407 12.7708 14.5567 12.5564C14.8775 12.0964 15.0656 11.5375 15.0656 10.9333C15.0656 10.3291 14.8775 9.77025 14.5567 9.31028C14.407 9.09581 14.4596 8.80066 14.6741 8.65106Z" fill="#444CE7"/>
+<path fill-rule="evenodd" clip-rule="evenodd" d="M12.5674 6.53771C12.794 6.51987 13.0155 6.61161 13.1632 6.78449C13.2954 6.93929 13.3164 7.12549 13.3244 7.21587C13.3334 7.31718 13.3334 7.44301 13.3333 7.57103C13.3333 7.57691 13.3333 7.58278 13.3333 7.58866L13.3333 14.3C13.3334 14.428 13.3334 14.5539 13.3244 14.6552C13.3164 14.7455 13.2954 14.9317 13.1632 15.0865C13.0155 15.2594 12.794 15.3512 12.5674 15.3333C12.3644 15.3173 12.2179 15.2005 12.1484 15.1423C12.0704 15.077 11.9814 14.988 11.8909 14.8975L10.3795 13.3861C10.3357 13.3423 10.3137 13.3205 10.2971 13.3053L10.2958 13.3041L10.2941 13.3041C10.2716 13.303 10.2407 13.3029 10.1787 13.3029L9.34101 13.3029C9.22151 13.3029 9.10513 13.3029 9.00657 13.2949C8.89833 13.286 8.77062 13.2652 8.6421 13.1997C8.46392 13.1089 8.31906 12.964 8.22827 12.7859C8.16279 12.6574 8.14192 12.5296 8.13308 12.4214C8.12503 12.3228 8.12504 12.2065 8.12505 12.087V9.79916C8.12505 9.79413 8.12505 9.78909 8.12505 9.78406C8.12504 9.66456 8.12503 9.54819 8.13308 9.44963C8.14192 9.34139 8.16279 9.21368 8.22827 9.08517C8.31906 8.90699 8.46392 8.76212 8.6421 8.67133C8.77062 8.60585 8.89833 8.58498 9.00657 8.57614C9.10512 8.56809 9.2215 8.5681 9.341 8.56812C9.34603 8.56812 9.35106 8.56812 9.3561 8.56812H10.1787C10.2407 8.56812 10.2716 8.56801 10.2941 8.56698L10.2958 8.5669L10.2971 8.56575C10.3137 8.55058 10.3357 8.52877 10.3795 8.48491L11.8784 6.98602C11.8826 6.98186 11.8867 6.97771 11.8909 6.97355C11.9814 6.88302 12.0704 6.79403 12.1484 6.72874C12.2179 6.67049 12.3644 6.55368 12.5674 6.53771Z" fill="#444CE7"/>
+</g>
+<defs>
+<clipPath id="clip0_109_6694">
+<rect width="16" height="16" fill="white"/>
+</clipPath>
+</defs>
+</svg>

web/app/components/base/icons/src/vender/line/mediaAndDevices/Speaker.json

+{
+	"icon": {
+		"type": "element",
+		"isRootNode": true,
+		"name": "svg",
+		"attributes": {
+			"width": "16",
+			"height": "16",
+			"viewBox": "0 0 16 16",
+			"fill": "none",
+			"xmlns": "http://www.w3.org/2000/svg"
+		},
+		"children": [
+			{
+				"type": "element",
+				"name": "g",
+				"attributes": {
+					"clip-path": "url(#clip0_109_6694)"
+				},
+				"children": [
+					{
+						"type": "element",
+						"name": "path",
+						"attributes": {
+							"fill-rule": "evenodd",
+							"clip-rule": "evenodd",
+							"d": "M0 2.86666C0 2.05664 0.656649 1.39999 1.46667 1.39999H5.86667C6.67668 1.39999 7.33333 2.05664 7.33333 2.86666C7.33333 3.27167 7.00501 3.59999 6.6 3.59999C6.19499 3.59999 5.86667 3.27167 5.86667 2.86666H4.4V7.99999C4.80501 7.99999 5.13333 8.32831 5.13333 8.73332C5.13333 9.13833 4.80501 9.46666 4.4 9.46666H2.93333C2.52832 9.46666 2.2 9.13833 2.2 8.73332C2.2 8.32831 2.52832 7.99999 2.93333 7.99999V2.86666H1.46667C1.46667 3.27167 1.13834 3.59999 0.733333 3.59999C0.328324 3.59999 0 3.27167 0 2.86666Z",
+							"fill": "currentColor"
+						},
+						"children": []
+					},
+					{
+						"type": "element",
+						"name": "path",
+						"attributes": {
+							"d": "M13.8205 0.782296C13.7434 0.62811 13.5233 0.62811 13.4462 0.782296C12.9664 1.74206 12.8754 1.83302 11.9156 2.3129C11.7615 2.39 11.7615 2.61003 11.9156 2.68712C12.8754 3.167 12.9664 3.25797 13.4462 4.21773C13.5233 4.37191 13.7434 4.37191 13.8205 4.21773C14.3003 3.25797 14.3913 3.167 15.3511 2.68712C15.5053 2.61003 15.5053 2.39 15.3511 2.3129C14.3913 1.83302 14.3003 1.74206 13.8205 0.782296Z",
+							"fill": "currentColor"
+						},
+						"children": []
+					},
+					{
+						"type": "element",
+						"name": "path",
+						"attributes": {
+							"d": "M9.79394 2.25319C9.71404 2.09337 9.48596 2.09337 9.40605 2.25319C9.04994 2.96543 8.96544 3.04993 8.2532 3.40605C8.09338 3.48595 8.09338 3.71402 8.2532 3.79393C8.96544 4.15005 9.04994 4.23455 9.40606 4.94679C9.48596 5.10661 9.71404 5.10661 9.79394 4.94679C10.1501 4.23455 10.2346 4.15005 10.9468 3.79393C11.1066 3.71402 11.1066 3.48595 10.9468 3.40605C10.2346 3.04993 10.1501 2.96543 9.79394 2.25319Z",
+							"fill": "currentColor"
+						},
+						"children": []
+					},
+					{
+						"type": "element",
+						"name": "path",
+						"attributes": {
+							"d": "M2.75377 11.049C2.67668 10.8948 2.45665 10.8948 2.37956 11.049C1.89969 12.0087 1.80872 12.0997 0.848971 12.5796C0.694788 12.6566 0.694787 12.8767 0.848971 12.9538C1.80872 13.4336 1.89969 13.5246 2.37956 14.4844C2.45665 14.6385 2.67668 14.6385 2.75377 14.4844C3.23365 13.5246 3.32461 13.4336 4.28436 12.9538C4.43855 12.8767 4.43855 12.6566 4.28436 12.5796C3.32461 12.0997 3.23365 12.0087 2.75377 11.049Z",
+							"fill": "currentColor"
+						},
+						"children": []
+					},
+					{
+						"type": "element",
+						"name": "path",
+						"attributes": {
+							"fill-rule": "evenodd",
+							"clip-rule": "evenodd",
+							"d": "M14.6741 8.65106C14.8886 8.50146 15.1837 8.55405 15.3333 8.76853C15.7614 9.38226 16.0125 10.1292 16.0125 10.9333C16.0125 11.7375 15.7614 12.4844 15.3333 13.0981C15.1837 13.3126 14.8886 13.3652 14.6741 13.2156C14.4596 13.066 14.407 12.7708 14.5567 12.5564C14.8775 12.0964 15.0656 11.5375 15.0656 10.9333C15.0656 10.3291 14.8775 9.77025 14.5567 9.31028C14.407 9.09581 14.4596 8.80066 14.6741 8.65106Z",
+							"fill": "currentColor"
+						},
+						"children": []
+					},
+					{
+						"type": "element",
+						"name": "path",
+						"attributes": {
+							"fill-rule": "evenodd",
+							"clip-rule": "evenodd",
+							"d": "M12.5674 6.53771C12.794 6.51987 13.0155 6.61161 13.1632 6.78449C13.2954 6.93929 13.3164 7.12549 13.3244 7.21587C13.3334 7.31718 13.3334 7.44301 13.3333 7.57103C13.3333 7.57691 13.3333 7.58278 13.3333 7.58866L13.3333 14.3C13.3334 14.428 13.3334 14.5539 13.3244 14.6552C13.3164 14.7455 13.2954 14.9317 13.1632 15.0865C13.0155 15.2594 12.794 15.3512 12.5674 15.3333C12.3644 15.3173 12.2179 15.2005 12.1484 15.1423C12.0704 15.077 11.9814 14.988 11.8909 14.8975L10.3795 13.3861C10.3357 13.3423 10.3137 13.3205 10.2971 13.3053L10.2958 13.3041L10.2941 13.3041C10.2716 13.303 10.2407 13.3029 10.1787 13.3029L9.34101 13.3029C9.22151 13.3029 9.10513 13.3029 9.00657 13.2949C8.89833 13.286 8.77062 13.2652 8.6421 13.1997C8.46392 13.1089 8.31906 12.964 8.22827 12.7859C8.16279 12.6574 8.14192 12.5296 8.13308 12.4214C8.12503 12.3228 8.12504 12.2065 8.12505 12.087V9.79916C8.12505 9.79413 8.12505 9.78909 8.12505 9.78406C8.12504 9.66456 8.12503 9.54819 8.13308 9.44963C8.14192 9.34139 8.16279 9.21368 8.22827 9.08517C8.31906 8.90699 8.46392 8.76212 8.6421 8.67133C8.77062 8.60585 8.89833 8.58498 9.00657 8.57614C9.10512 8.56809 9.2215 8.5681 9.341 8.56812C9.34603 8.56812 9.35106 8.56812 9.3561 8.56812H10.1787C10.2407 8.56812 10.2716 8.56801 10.2941 8.56698L10.2958 8.5669L10.2971 8.56575C10.3137 8.55058 10.3357 8.52877 10.3795 8.48491L11.8784 6.98602C11.8826 6.98186 11.8867 6.97771 11.8909 6.97355C11.9814 6.88302 12.0704 6.79403 12.1484 6.72874C12.2179 6.67049 12.3644 6.55368 12.5674 6.53771Z",
+							"fill": "currentColor"
+						},
+						"children": []
+					}
+				]
+			},
+			{
+				"type": "element",
+				"name": "defs",
+				"attributes": {},
+				"children": [
+					{
+						"type": "element",
+						"name": "clipPath",
+						"attributes": {
+							"id": "clip0_109_6694"
+						},
+						"children": [
+							{
+								"type": "element",
+								"name": "rect",
+								"attributes": {
+									"width": "16",
+									"height": "16",
+									"fill": "white"
+								},
+								"children": []
+							}
+						]
+					}
+				]
+			}
+		]
+	},
+	"name": "Speaker"
+}
\ No newline at end of file

web/app/components/base/icons/src/vender/line/mediaAndDevices/Speaker.tsx

+// GENERATE BY script
+// DON NOT EDIT IT MANUALLY
+
+import * as React from 'react'
+import data from './Speaker.json'
+import IconBase from '@/app/components/base/icons/IconBase'
+import type { IconBaseProps, IconData } from '@/app/components/base/icons/IconBase'
+
+const Icon = React.forwardRef<React.MutableRefObject<SVGElement>, Omit<IconBaseProps, 'data'>>((
+  props,
+  ref,
+) => <IconBase {...props} ref={ref} data={data as IconData} />)
+
+Icon.displayName = 'Speaker'
+
+export default Icon

web/app/components/base/icons/src/vender/line/mediaAndDevices/index.ts

 export { default as Microphone01 } from './Microphone01'
 export { default as SlidersH } from './SlidersH'
+export { default as Speaker } from './Speaker'

web/app/components/base/icons/src/vender/solid/mediaAndDevices/Speaker.json

+{
+	"icon": {
+		"type": "element",
+		"isRootNode": true,
+		"name": "svg",
+		"attributes": {
+			"width": "16",
+			"height": "16",
+			"viewBox": "0 0 16 16",
+			"fill": "none",
+			"xmlns": "http://www.w3.org/2000/svg"
+		},
+		"children": [
+			{
+				"type": "element",
+				"name": "g",
+				"attributes": {
+					"clip-path": "url(#clip0_109_6694)"
+				},
+				"children": [
+					{
+						"type": "element",
+						"name": "path",
+						"attributes": {
+							"fill-rule": "evenodd",
+							"clip-rule": "evenodd",
+							"d": "M0 2.86666C0 2.05664 0.656649 1.39999 1.46667 1.39999H5.86667C6.67668 1.39999 7.33333 2.05664 7.33333 2.86666C7.33333 3.27167 7.00501 3.59999 6.6 3.59999C6.19499 3.59999 5.86667 3.27167 5.86667 2.86666H4.4V7.99999C4.80501 7.99999 5.13333 8.32831 5.13333 8.73332C5.13333 9.13833 4.80501 9.46666 4.4 9.46666H2.93333C2.52832 9.46666 2.2 9.13833 2.2 8.73332C2.2 8.32831 2.52832 7.99999 2.93333 7.99999V2.86666H1.46667C1.46667 3.27167 1.13834 3.59999 0.733333 3.59999C0.328324 3.59999 0 3.27167 0 2.86666Z",
+							"fill": "currentColor"
+						},
+						"children": []
+					},
+					{
+						"type": "element",
+						"name": "path",
+						"attributes": {
+							"d": "M13.8205 0.782296C13.7434 0.62811 13.5233 0.62811 13.4462 0.782296C12.9664 1.74206 12.8754 1.83302 11.9156 2.3129C11.7615 2.39 11.7615 2.61003 11.9156 2.68712C12.8754 3.167 12.9664 3.25797 13.4462 4.21773C13.5233 4.37191 13.7434 4.37191 13.8205 4.21773C14.3003 3.25797 14.3913 3.167 15.3511 2.68712C15.5053 2.61003 15.5053 2.39 15.3511 2.3129C14.3913 1.83302 14.3003 1.74206 13.8205 0.782296Z",
+							"fill": "currentColor"
+						},
+						"children": []
+					},
+					{
+						"type": "element",
+						"name": "path",
+						"attributes": {
+							"d": "M9.79394 2.25319C9.71404 2.09337 9.48596 2.09337 9.40605 2.25319C9.04994 2.96543 8.96544 3.04993 8.2532 3.40605C8.09338 3.48595 8.09338 3.71402 8.2532 3.79393C8.96544 4.15005 9.04994 4.23455 9.40606 4.94679C9.48596 5.10661 9.71404 5.10661 9.79394 4.94679C10.1501 4.23455 10.2346 4.15005 10.9468 3.79393C11.1066 3.71402 11.1066 3.48595 10.9468 3.40605C10.2346 3.04993 10.1501 2.96543 9.79394 2.25319Z",
+							"fill": "currentColor"
+						},
+						"children": []
+					},
+					{
+						"type": "element",
+						"name": "path",
+						"attributes": {
+							"d": "M2.75377 11.049C2.67668 10.8948 2.45665 10.8948 2.37956 11.049C1.89969 12.0087 1.80872 12.0997 0.848971 12.5796C0.694788 12.6566 0.694787 12.8767 0.848971 12.9538C1.80872 13.4336 1.89969 13.5246 2.37956 14.4844C2.45665 14.6385 2.67668 14.6385 2.75377 14.4844C3.23365 13.5246 3.32461 13.4336 4.28436 12.9538C4.43855 12.8767 4.43855 12.6566 4.28436 12.5796C3.32461 12.0997 3.23365 12.0087 2.75377 11.049Z",
+							"fill": "currentColor"
+						},
+						"children": []
+					},
+					{
+						"type": "element",
+						"name": "path",
+						"attributes": {
+							"fill-rule": "evenodd",
+							"clip-rule": "evenodd",
+							"d": "M14.6741 8.65106C14.8886 8.50146 15.1837 8.55405 15.3333 8.76853C15.7614 9.38226 16.0125 10.1292 16.0125 10.9333C16.0125 11.7375 15.7614 12.4844 15.3333 13.0981C15.1837 13.3126 14.8886 13.3652 14.6741 13.2156C14.4596 13.066 14.407 12.7708 14.5567 12.5564C14.8775 12.0964 15.0656 11.5375 15.0656 10.9333C15.0656 10.3291 14.8775 9.77025 14.5567 9.31028C14.407 9.09581 14.4596 8.80066 14.6741 8.65106Z",
+							"fill": "currentColor"
+						},
+						"children": []
+					},
+					{
+						"type": "element",
+						"name": "path",
+						"attributes": {
+							"fill-rule": "evenodd",
+							"clip-rule": "evenodd",
+							"d": "M12.5674 6.53771C12.794 6.51987 13.0155 6.61161 13.1632 6.78449C13.2954 6.93929 13.3164 7.12549 13.3244 7.21587C13.3334 7.31718 13.3334 7.44301 13.3333 7.57103C13.3333 7.57691 13.3333 7.58278 13.3333 7.58866L13.3333 14.3C13.3334 14.428 13.3334 14.5539 13.3244 14.6552C13.3164 14.7455 13.2954 14.9317 13.1632 15.0865C13.0155 15.2594 12.794 15.3512 12.5674 15.3333C12.3644 15.3173 12.2179 15.2005 12.1484 15.1423C12.0704 15.077 11.9814 14.988 11.8909 14.8975L10.3795 13.3861C10.3357 13.3423 10.3137 13.3205 10.2971 13.3053L10.2958 13.3041L10.2941 13.3041C10.2716 13.303 10.2407 13.3029 10.1787 13.3029L9.34101 13.3029C9.22151 13.3029 9.10513 13.3029 9.00657 13.2949C8.89833 13.286 8.77062 13.2652 8.6421 13.1997C8.46392 13.1089 8.31906 12.964 8.22827 12.7859C8.16279 12.6574 8.14192 12.5296 8.13308 12.4214C8.12503 12.3228 8.12504 12.2065 8.12505 12.087V9.79916C8.12505 9.79413 8.12505 9.78909 8.12505 9.78406C8.12504 9.66456 8.12503 9.54819 8.13308 9.44963C8.14192 9.34139 8.16279 9.21368 8.22827 9.08517C8.31906 8.90699 8.46392 8.76212 8.6421 8.67133C8.77062 8.60585 8.89833 8.58498 9.00657 8.57614C9.10512 8.56809 9.2215 8.5681 9.341 8.56812C9.34603 8.56812 9.35106 8.56812 9.3561 8.56812H10.1787C10.2407 8.56812 10.2716 8.56801 10.2941 8.56698L10.2958 8.5669L10.2971 8.56575C10.3137 8.55058 10.3357 8.52877 10.3795 8.48491L11.8784 6.98602C11.8826 6.98186 11.8867 6.97771 11.8909 6.97355C11.9814 6.88302 12.0704 6.79403 12.1484 6.72874C12.2179 6.67049 12.3644 6.55368 12.5674 6.53771Z",
+							"fill": "currentColor"
+						},
+						"children": []
+					}
+				]
+			},
+			{
+				"type": "element",
+				"name": "defs",
+				"attributes": {},
+				"children": [
+					{
+						"type": "element",
+						"name": "clipPath",
+						"attributes": {
+							"id": "clip0_109_6694"
+						},
+						"children": [
+							{
+								"type": "element",
+								"name": "rect",
+								"attributes": {
+									"width": "16",
+									"height": "16",
+									"fill": "white"
+								},
+								"children": []
+							}
+						]
+					}
+				]
+			}
+		]
+	},
+	"name": "Speaker"
+}
\ No newline at end of file

web/app/components/base/icons/src/vender/solid/mediaAndDevices/Speaker.tsx

+// GENERATE BY script
+// DON NOT EDIT IT MANUALLY
+
+import * as React from 'react'
+import data from './Speaker.json'
+import IconBase from '@/app/components/base/icons/IconBase'
+import type { IconBaseProps, IconData } from '@/app/components/base/icons/IconBase'
+
+const Icon = React.forwardRef<React.MutableRefObject<SVGElement>, Omit<IconBaseProps, 'data'>>((
+  props,
+  ref,
+) => <IconBase {...props} ref={ref} data={data as IconData} />)
+
+Icon.displayName = 'Speaker'
+
+export default Icon

web/app/components/base/icons/src/vender/solid/mediaAndDevices/index.ts

@@ -4,4 +4,5 @@ export { default as MagicWand } from './MagicWand'
 export { default as Microphone01 } from './Microphone01'
 export { default as Robot } from './Robot'
 export { default as Sliders02 } from './Sliders02'
+export { default as Speaker } from './Speaker'
 export { default as StopCircle } from './StopCircle'

web/app/components/develop/secret-key/assets/pause.svg

+<svg width="24" height="24" viewBox="0 0 24 24" fill="none" xmlns="http://www.w3.org/2000/svg">
+<g clip-path="url(#clip0_129_2189)">
+<path d="M10.6666 14V10M13.3333 14V10M18.6666 12C18.6666 15.6819 15.6819 18.6667 12 18.6667C8.31808 18.6667 5.33331 15.6819 5.33331 12C5.33331 8.3181 8.31808 5.33333 12 5.33333C15.6819 5.33333 18.6666 8.3181 18.6666 12Z" stroke="#155EEF" stroke-width="1.5" stroke-linecap="round" stroke-linejoin="round"/>
+</g>
+<defs>
+<clipPath id="clip0_129_2189">
+<rect width="16" height="16" fill="white" transform="translate(4 4)"/>
+</clipPath>
+</defs>
+</svg>

web/app/components/develop/secret-key/assets/play.svg

+<svg width="16" height="16" viewBox="0 0 16 16" fill="none" xmlns="http://www.w3.org/2000/svg">
+<g clip-path="url(#clip0_129_107)">
+<path d="M7.99991 14.6666C11.6819 14.6666 14.6666 11.6819 14.6666 7.99998C14.6666 4.31808 11.6819 1.33331 7.99998 1.33331C4.31808 1.33331 1.33331 4.31808 1.33331 7.99998C1.33331 11.6819 4.31808 14.6666 7.99998 14.6666Z" stroke="#155EEF" stroke-width="1.5" stroke-linecap="round" stroke-linejoin="round"/>
+<path d="M6.66665 5.33331L10.6666 7.99998L6.66665 10.6666V5.33331Z" stroke="#155EEF" stroke-width="1.5" stroke-linecap="round" stroke-linejoin="round"/>
+</g>
+<defs>
+<clipPath id="clip0_129_107">
+<rect width="16" height="16" fill="white"/>
+</clipPath>
+</defs>
+</svg>

web/app/components/develop/secret-key/assets/stop.svg

+<svg width="16" height="16" viewBox="0 0 16 16" fill="none" xmlns="http://www.w3.org/2000/svg">
+<g clip-path="url(#clip0_129_107)">
+<path d="M7.99998 14.6666C11.6819 14.6666 14.6666 11.6819 14.6666 7.99998C14.6666 4.31808 11.6819 1.33331 7.99998 1.33331C4.31808 1.33331 1.33331 4.31808 1.33331 7.99998C1.33331 11.6819 4.31808 14.6666 7.99998 14.6666Z" stroke="#667085" stroke-width="1.5" stroke-linecap="round" stroke-linejoin="round"/>
+<path d="M6.66665 5.33331L10.6666 7.99998L6.66665 10.6666V5.33331Z" stroke="#667085" stroke-width="1.5" stroke-linecap="round" stroke-linejoin="round"/>
+</g>
+<defs>
+<clipPath id="clip0_129_107">
+<rect width="16" height="16" fill="white"/>
+</clipPath>
+</defs>
+</svg>

web/app/components/develop/template/template.en.mdx

@@ -484,3 +484,51 @@ The text generation application offers non-session support and is ideal for tran
     </CodeGroup>
   </Col>
 </Row>
+
+---
+
+<Heading
+  url='/text-to-audio'
+  method='POST'
+  title='text to audio'
+  name='#audio'
+/>
+<Row>
+  <Col>
+    Text to speech, only supports openai model.
+
+    ### Request Body
+
+    <Properties>
+      <Property name='text' type='str' key='text'>
+        Speech generated content。
+      </Property>
+      <Property name='user' type='string' key='user'>
+        The user identifier, defined by the developer, must ensure uniqueness within the app.
+      </Property>
+      <Property name='streaming' type='bool' key='streaming'>
+        Whether to enable streaming output, true、false。
+      </Property>
+    </Properties>
+  </Col>
+  <Col sticky>
+
+    <CodeGroup title="Request" tag="POST" label="/text-to-audio" targetCode={`curl --location --request POST '${props.appDetail.api_base_url}/text-to-audio' \\\n--header 'Authorization: Bearer ENTER-YOUR-SECRET-KEY' \\\n--form 'text=Hello Dify;user=abc-123;streaming=false`}>
+
+    ```bash {{ title: 'cURL' }}
+    curl --location --request POST '${props.appDetail.api_base_url}/text-to-audio' \
+    --header 'Authorization: Bearer ENTER-YOUR-SECRET-KEY' \
+    --form 'file=Hello Dify;user=abc-123;streaming=false'
+    ```
+
+    </CodeGroup>
+
+    <CodeGroup title="headers">
+    ```json {{ title: 'headers' }}
+    {
+      "Content-Type": "audio/wav"
+    }
+    ```
+    </CodeGroup>
+  </Col>
+</Row>
\ No newline at end of file

web/app/components/develop/template/template.zh.mdx

@@ -447,3 +447,51 @@ import { Row, Col, Properties, Property, Heading, SubProperty } from '../md.tsx'
     </CodeGroup>
   </Col>
 </Row>
+
+---
+
+<Heading
+  url='/text-to-audio'
+  method='POST'
+  title='文字转语音'
+  name='#audio'
+/>
+<Row>
+  <Col>
+    文字转语音，仅支持 openai 模型。
+
+    ### Request Body
+
+    <Properties>
+      <Property name='text' type='str' key='text'>
+        语音生成内容。
+      </Property>
+      <Property name='user' type='string' key='user'>
+        用户标识，由开发者定义规则，需保证用户标识在应用内唯一。
+      </Property>
+      <Property name='streaming' type='bool' key='streaming'>
+        是否启用流式输出true、false。
+      </Property>
+    </Properties>
+  </Col>
+  <Col sticky>
+
+    <CodeGroup title="Request" tag="POST" label="/text-to-audio" targetCode={`curl --location --request POST '${props.appDetail.api_base_url}/text-to-audio' \\\n--header 'Authorization: Bearer ENTER-YOUR-SECRET-KEY' \\\n--form 'text=你好Dify;user=abc-123;streaming=false`}>
+
+    ```bash {{ title: 'cURL' }}
+    curl --location --request POST '${props.appDetail.api_base_url}/text-to-audio' \
+    --header 'Authorization: Bearer ENTER-YOUR-SECRET-KEY' \
+    --form 'file=你好Dify;user=abc-123;streaming=false'
+    ```
+
+    </CodeGroup>
+
+    <CodeGroup title="headers">
+    ```json {{ title: 'headers' }}
+    {
+      "Content-Type": "audio/wav"
+    }
+    ```
+    </CodeGroup>
+  </Col>
+</Row>

web/app/components/develop/template/template_chat.en.mdx

@@ -837,6 +837,54 @@ Chat applications support session persistence, allowing previous chat history to
 
 ---
 
+<Heading
+  url='/text-to-audio'
+  method='POST'
+  title='text to audio'
+  name='#audio'
+/>
+<Row>
+  <Col>
+    Text to speech, only supports openai model.
+
+    ### Request Body
+
+    <Properties>
+      <Property name='text' type='str' key='text'>
+        Speech generated content。
+      </Property>
+      <Property name='user' type='string' key='user'>
+        The user identifier, defined by the developer, must ensure uniqueness within the app.
+      </Property>
+      <Property name='streaming' type='bool' key='streaming'>
+        Whether to enable streaming output, true、false。
+      </Property>
+    </Properties>
+  </Col>
+  <Col sticky>
+
+    <CodeGroup title="Request" tag="POST" label="/text-to-audio" targetCode={`curl --location --request POST '${props.appDetail.api_base_url}/text-to-audio' \\\n--header 'Authorization: Bearer ENTER-YOUR-SECRET-KEY' \\\n--form 'text=Hello Dify;user=abc-123;streaming=false`}>
+
+    ```bash {{ title: 'cURL' }}
+    curl --location --request POST '${props.appDetail.api_base_url}/text-to-audio' \
+    --header 'Authorization: Bearer ENTER-YOUR-SECRET-KEY' \
+    --form 'file=Hello Dify;user=abc-123;streaming=false'
+    ```
+
+    </CodeGroup>
+
+    <CodeGroup title="headers">
+    ```json {{ title: 'headers' }}
+    {
+      "Content-Type": "audio/wav"
+    }
+    ```
+    </CodeGroup>
+  </Col>
+</Row>
+
+---
+
 <Heading
   url='/parameters'
   method='GET'

web/app/components/develop/template/template_chat.zh.mdx

@@ -909,6 +909,54 @@ import { Row, Col, Properties, Property, Heading, SubProperty } from '../md.tsx'
 
 ---
 
+<Heading
+  url='/text-to-audio'
+  method='POST'
+  title='文字转语音'
+  name='#audio'
+/>
+<Row>
+  <Col>
+    文字转语音，仅支持 openai 模型。
+
+    ### Request Body
+
+    <Properties>
+      <Property name='text' type='str' key='text'>
+        语音生成内容。
+      </Property>
+      <Property name='user' type='string' key='user'>
+        用户标识，由开发者定义规则，需保证用户标识在应用内唯一。
+      </Property>
+      <Property name='streaming' type='bool' key='streaming'>
+        是否启用流式输出true、false。
+      </Property>
+    </Properties>
+  </Col>
+  <Col sticky>
+
+    <CodeGroup title="Request" tag="POST" label="/text-to-audio" targetCode={`curl --location --request POST '${props.appDetail.api_base_url}/text-to-audio' \\\n--header 'Authorization: Bearer ENTER-YOUR-SECRET-KEY' \\\n--form 'text=你好Dify;user=abc-123;streaming=false`}>
+
+    ```bash {{ title: 'cURL' }}
+    curl --location --request POST '${props.appDetail.api_base_url}/text-to-audio' \
+    --header 'Authorization: Bearer ENTER-YOUR-SECRET-KEY' \
+    --form 'file=你好Dify;user=abc-123;streaming=false'
+    ```
+
+    </CodeGroup>
+
+    <CodeGroup title="headers">
+    ```json {{ title: 'headers' }}
+    {
+      "Content-Type": "audio/wav"
+    }
+    ```
+    </CodeGroup>
+  </Col>
+</Row>
+
+---
+
 <Heading
   url='/parameters'
   method='GET'

web/app/components/header/account-setting/model-provider-page/declarations.ts

@@ -26,6 +26,7 @@ export enum ModelTypeEnum {
   rerank = 'rerank',
   speech2text = 'speech2text',
   moderation = 'moderation',
+  tts = 'tts',
 }
 
 export const MODEL_TYPE_TEXT = {
@@ -34,6 +35,7 @@ export const MODEL_TYPE_TEXT = {
   [ModelTypeEnum.rerank]: 'Rerank',
   [ModelTypeEnum.speech2text]: 'Speech2text',
   [ModelTypeEnum.moderation]: 'Moderation',
+  [ModelTypeEnum.tts]: 'TTS',
 }
 
 export enum ConfigurateMethodEnum {

web/app/components/header/account-setting/model-provider-page/hooks.ts

@@ -100,12 +100,13 @@ export const useProviderCrenditialsFormSchemasValue = (
   return value
 }
 
-export type ModelTypeIndex = 1 | 2 | 3 | 4
+export type ModelTypeIndex = 1 | 2 | 3 | 4 | 5
 export const MODEL_TYPE_MAPS = {
   1: ModelTypeEnum.textGeneration,
   2: ModelTypeEnum.textEmbedding,
   3: ModelTypeEnum.rerank,
   4: ModelTypeEnum.speech2text,
+  5: ModelTypeEnum.tts,
 }
 
 export const useModelList = (type: ModelTypeIndex) => {

web/app/components/header/account-setting/model-provider-page/index.tsx

@@ -30,9 +30,10 @@ const ModelProviderPage = () => {
   const { data: embeddingsDefaultModel } = useDefaultModel(2)
   const { data: rerankDefaultModel } = useDefaultModel(3)
   const { data: speech2textDefaultModel } = useDefaultModel(4)
+  const { data: ttsDefaultModel } = useDefaultModel(5)
   const { modelProviders: providers } = useProviderContext()
   const { setShowModelModal } = useModalContext()
-  const defaultModelNotConfigured = !textGenerationDefaultModel && !embeddingsDefaultModel && !speech2textDefaultModel && !rerankDefaultModel
+  const defaultModelNotConfigured = !textGenerationDefaultModel && !embeddingsDefaultModel && !speech2textDefaultModel && !rerankDefaultModel && !ttsDefaultModel
   const [configedProviders, notConfigedProviders] = useMemo(() => {
     const configedProviders: ModelProvider[] = []
     const notConfigedProviders: ModelProvider[] = []
@@ -104,6 +105,7 @@ const ModelProviderPage = () => {
           embeddingsDefaultModel={embeddingsDefaultModel}
           rerankDefaultModel={rerankDefaultModel}
           speech2textDefaultModel={speech2textDefaultModel}
+          ttsDefaultModel={ttsDefaultModel}
         />
       </div>
       {

web/app/components/header/account-setting/model-provider-page/system-model-selector/index.tsx

@@ -29,12 +29,14 @@ type SystemModelSelectorProps = {
   embeddingsDefaultModel: DefaultModelResponse | undefined
   rerankDefaultModel: DefaultModelResponse | undefined
   speech2textDefaultModel: DefaultModelResponse | undefined
+  ttsDefaultModel: DefaultModelResponse | undefined
 }
 const SystemModel: FC<SystemModelSelectorProps> = ({
   textGenerationDefaultModel,
   embeddingsDefaultModel,
   rerankDefaultModel,
   speech2textDefaultModel,
+  ttsDefaultModel,
 }) => {
   const { t } = useTranslation()
   const { notify } = useToastContext()
@@ -43,11 +45,13 @@ const SystemModel: FC<SystemModelSelectorProps> = ({
   const { data: embeddingModelList } = useModelList(2)
   const { data: rerankModelList } = useModelList(3)
   const { data: speech2textModelList } = useModelList(4)
+  const { data: ttsModelList } = useModelList(5)
   const [changedModelTypes, setChangedModelTypes] = useState<ModelTypeEnum[]>([])
   const [currentTextGenerationDefaultModel, changeCurrentTextGenerationDefaultModel] = useSystemDefaultModelAndModelList(textGenerationDefaultModel, textGenerationModelList)
   const [currentEmbeddingsDefaultModel, changeCurrentEmbeddingsDefaultModel] = useSystemDefaultModelAndModelList(embeddingsDefaultModel, embeddingModelList)
   const [currentRerankDefaultModel, changeCurrentRerankDefaultModel] = useSystemDefaultModelAndModelList(rerankDefaultModel, rerankModelList)
   const [currentSpeech2textDefaultModel, changeCurrentSpeech2textDefaultModel] = useSystemDefaultModelAndModelList(speech2textDefaultModel, speech2textModelList)
+  const [currentTTSDefaultModel, changeCurrentTTSDefaultModel] = useSystemDefaultModelAndModelList(ttsDefaultModel, ttsModelList)
   const [open, setOpen] = useState(false)
 
   const getCurrentDefaultModelByModelType = (modelType: ModelTypeEnum) => {
@@ -59,6 +63,8 @@ const SystemModel: FC<SystemModelSelectorProps> = ({
       return currentRerankDefaultModel
     else if (modelType === ModelTypeEnum.speech2text)
       return currentSpeech2textDefaultModel
+    else if (modelType === ModelTypeEnum.tts)
+      return currentTTSDefaultModel
 
     return undefined
   }
@@ -71,6 +77,8 @@ const SystemModel: FC<SystemModelSelectorProps> = ({
       changeCurrentRerankDefaultModel(model)
     else if (modelType === ModelTypeEnum.speech2text)
       changeCurrentSpeech2textDefaultModel(model)
+    else if (modelType === ModelTypeEnum.tts)
+      changeCurrentTTSDefaultModel(model)
 
     if (!changedModelTypes.includes(modelType))
       setChangedModelTypes([...changedModelTypes, modelType])
@@ -79,7 +87,7 @@ const SystemModel: FC<SystemModelSelectorProps> = ({
     const res = await updateDefaultModel({
       url: '/workspaces/current/default-model',
       body: {
-        model_settings: [ModelTypeEnum.textGeneration, ModelTypeEnum.textEmbedding, ModelTypeEnum.rerank, ModelTypeEnum.speech2text].map((modelType) => {
+        model_settings: [ModelTypeEnum.textGeneration, ModelTypeEnum.textEmbedding, ModelTypeEnum.rerank, ModelTypeEnum.speech2text, ModelTypeEnum.tts].map((modelType) => {
           return {
             model_type: modelType,
             provider: getCurrentDefaultModelByModelType(modelType)?.provider,
@@ -101,6 +109,8 @@ const SystemModel: FC<SystemModelSelectorProps> = ({
           updateModelList(modelType)
         else if (modelType === ModelTypeEnum.speech2text)
           updateModelList(modelType)
+        else if (modelType === ModelTypeEnum.tts)
+          updateModelList(modelType)
       })
     }
   }
@@ -136,7 +146,7 @@ const SystemModel: FC<SystemModelSelectorProps> = ({
                   <div className='w-[261px] text-gray-500'>{t('common.modelProvider.systemReasoningModel.tip')}</div>
                 }
               >
-                <HelpCircle className='ml-0.5 w-[14px] h-[14px] text-gray-400' />
+                <HelpCircle className='ml-0.5 w-[14px] h-[14px] text-gray-400'/>
               </Tooltip>
             </div>
             <div>
@@ -156,7 +166,7 @@ const SystemModel: FC<SystemModelSelectorProps> = ({
                   <div className='w-[261px] text-gray-500'>{t('common.modelProvider.embeddingModel.tip')}</div>
                 }
               >
-                <HelpCircle className='ml-0.5 w-[14px] h-[14px] text-gray-400' />
+                <HelpCircle className='ml-0.5 w-[14px] h-[14px] text-gray-400'/>
               </Tooltip>
             </div>
             <div>
@@ -176,7 +186,7 @@ const SystemModel: FC<SystemModelSelectorProps> = ({
                   <div className='w-[261px] text-gray-500'>{t('common.modelProvider.rerankModel.tip')}</div>
                 }
               >
-                <HelpCircle className='ml-0.5 w-[14px] h-[14px] text-gray-400' />
+                <HelpCircle className='ml-0.5 w-[14px] h-[14px] text-gray-400'/>
               </Tooltip>
             </div>
             <div>
@@ -196,7 +206,7 @@ const SystemModel: FC<SystemModelSelectorProps> = ({
                   <div className='w-[261px] text-gray-500'>{t('common.modelProvider.speechToTextModel.tip')}</div>
                 }
               >
-                <HelpCircle className='ml-0.5 w-[14px] h-[14px] text-gray-400' />
+                <HelpCircle className='ml-0.5 w-[14px] h-[14px] text-gray-400'/>
               </Tooltip>
             </div>
             <div>
@@ -207,6 +217,26 @@ const SystemModel: FC<SystemModelSelectorProps> = ({
               />
             </div>
           </div>
+          <div className='px-6 py-1'>
+            <div className='flex items-center h-8 text-[13px] font-medium text-gray-900'>
+              {t('common.modelProvider.ttsModel.key')}
+              <Tooltip
+                selector='model-page-system-tts-model-tip'
+                htmlContent={
+                  <div className='w-[261px] text-gray-500'>{t('common.modelProvider.ttsModel.tip')}</div>
+                }
+              >
+                <HelpCircle className='ml-0.5 w-[14px] h-[14px] text-gray-400'/>
+              </Tooltip>
+            </div>
+            <div>
+              <ModelSelector
+                defaultModel={currentTTSDefaultModel}
+                modelList={ttsModelList}
+                onSelect={model => handleChangeDefaultModel(ModelTypeEnum.tts, model)}
+              />
+            </div>
+          </div>
           <div className='flex items-center justify-end px-6 py-4'>
             <Button
               className='mr-2 !h-8 !text-[13px]'

web/app/components/share/chat/index.tsx

@@ -32,7 +32,14 @@ import {
   updateFeedback,
 } from '@/service/share'
 import type { AppMeta, ConversationItem, SiteInfo } from '@/models/share'
-import type { PromptConfig, SuggestedQuestionsAfterAnswerConfig } from '@/models/debug'
+
+import type {
+  CitationConfig,
+  PromptConfig,
+  SpeechToTextConfig,
+  SuggestedQuestionsAfterAnswerConfig,
+  TextToSpeechConfig,
+} from '@/models/debug'
 import type { Feedbacktype, IChatItem } from '@/app/components/app/chat/type'
 import Chat from '@/app/components/app/chat'
 import { changeLanguage } from '@/i18n/i18next-config'
@@ -177,8 +184,9 @@ const Main: FC<IMainProps> = ({
   }
 
   const [suggestedQuestionsAfterAnswerConfig, setSuggestedQuestionsAfterAnswerConfig] = useState<SuggestedQuestionsAfterAnswerConfig | null>(null)
-  const [speechToTextConfig, setSpeechToTextConfig] = useState<SuggestedQuestionsAfterAnswerConfig | null>(null)
-  const [citationConfig, setCitationConfig] = useState<SuggestedQuestionsAfterAnswerConfig | null>(null)
+  const [speechToTextConfig, setSpeechToTextConfig] = useState<SpeechToTextConfig | null>(null)
+  const [textToSpeechConfig, setTextToSpeechConfig] = useState<TextToSpeechConfig | null>(null)
+  const [citationConfig, setCitationConfig] = useState<CitationConfig | null>(null)
 
   const [conversationIdChangeBecauseOfNew, setConversationIdChangeBecauseOfNew, getConversationIdChangeBecauseOfNew] = useGetState(false)
   const [isChatStarted, { setTrue: setChatStarted, setFalse: setChatNotStarted }] = useBoolean(false)
@@ -335,6 +343,7 @@ const Main: FC<IMainProps> = ({
     const caculatedPromptVariables = inputs || currInputs || null
     if (caculatedIntroduction && caculatedPromptVariables)
       caculatedIntroduction = replaceStringWithValues(caculatedIntroduction, promptConfig?.prompt_variables || [], caculatedPromptVariables)
+
     const openstatement = {
       id: `${Date.now()}`,
       content: caculatedIntroduction,
@@ -394,7 +403,7 @@ const Main: FC<IMainProps> = ({
         const isNotNewConversation = allConversations.some(item => item.id === _conversationId)
         setAllConversationList(allConversations)
         // fetch new conversation info
-        const { user_input_form, opening_statement: introduction, suggested_questions, suggested_questions_after_answer, speech_to_text, retriever_resource, file_upload, sensitive_word_avoidance }: any = appParams
+        const { user_input_form, opening_statement: introduction, suggested_questions, suggested_questions_after_answer, speech_to_text, text_to_speech, retriever_resource, file_upload, sensitive_word_avoidance }: any = appParams
         setVisionConfig({
           ...file_upload.image,
           image_file_size_limit: appParams?.system_parameters?.image_file_size_limit,
@@ -416,6 +425,7 @@ const Main: FC<IMainProps> = ({
         } as PromptConfig)
         setSuggestedQuestionsAfterAnswerConfig(suggested_questions_after_answer)
         setSpeechToTextConfig(speech_to_text)
+        setTextToSpeechConfig(text_to_speech)
         setCitationConfig(retriever_resource)
 
         // setConversationList(conversations as ConversationItem[])
@@ -887,6 +897,7 @@ const Main: FC<IMainProps> = ({
                     isShowSuggestion={doShowSuggestion}
                     suggestionList={suggestedQuestions}
                     isShowSpeechToText={speechToTextConfig?.enabled}
+                    isShowTextToSpeech={textToSpeechConfig?.enabled}
                     isShowCitation={citationConfig?.enabled && isInstalledApp}
                     visionConfig={{
                       ...visionConfig,

web/app/components/share/chatbot/index.tsx

@@ -128,6 +128,8 @@ const Main: FC<IMainProps> = ({
   }
   const [suggestedQuestionsAfterAnswerConfig, setSuggestedQuestionsAfterAnswerConfig] = useState<SuggestedQuestionsAfterAnswerConfig | null>(null)
   const [speechToTextConfig, setSpeechToTextConfig] = useState<SuggestedQuestionsAfterAnswerConfig | null>(null)
+  const [textToSpeechConfig, setTextToSpeechConfig] = useState<SuggestedQuestionsAfterAnswerConfig | null>(null)
+  const [citationConfig, setCitationConfig] = useState<SuggestedQuestionsAfterAnswerConfig | null>(null)
 
   const [conversationIdChangeBecauseOfNew, setConversationIdChangeBecauseOfNew, getConversationIdChangeBecauseOfNew] = useGetState(false)
   const [isChatStarted, { setTrue: setChatStarted, setFalse: setChatNotStarted }] = useBoolean(false)
@@ -298,7 +300,7 @@ const Main: FC<IMainProps> = ({
         const isNotNewConversation = allConversations.some(item => item.id === _conversationId)
         setAllConversationList(allConversations)
         // fetch new conversation info
-        const { user_input_form, opening_statement: introduction, suggested_questions_after_answer, speech_to_text, file_upload, sensitive_word_avoidance }: any = appParams
+        const { user_input_form, opening_statement: introduction, suggested_questions_after_answer, speech_to_text, text_to_speech, retriever_resource, file_upload, sensitive_word_avoidance }: any = appParams
         setVisionConfig({
           ...file_upload.image,
           image_file_size_limit: appParams?.system_parameters?.image_file_size_limit,
@@ -318,6 +320,8 @@ const Main: FC<IMainProps> = ({
         } as PromptConfig)
         setSuggestedQuestionsAfterAnswerConfig(suggested_questions_after_answer)
         setSpeechToTextConfig(speech_to_text)
+        setTextToSpeechConfig(text_to_speech)
+        setCitationConfig(retriever_resource)
 
         // setConversationList(conversations as ConversationItem[])
 
@@ -634,6 +638,8 @@ const Main: FC<IMainProps> = ({
                     suggestionList={suggestQuestions}
                     displayScene='web'
                     isShowSpeechToText={speechToTextConfig?.enabled}
+                    isShowTextToSpeech={textToSpeechConfig?.enabled}
+                    isShowCitation={citationConfig?.enabled && isInstalledApp}
                     answerIcon={<LogoAvatar className='relative shrink-0' />}
                     visionConfig={visionConfig}
                   />

web/app/components/share/text-generation/index.tsx

@@ -16,7 +16,12 @@ import useBreakpoints, { MediaType } from '@/hooks/use-breakpoints'
 import RunOnce from '@/app/components/share/text-generation/run-once'
 import { fetchSavedMessage as doFetchSavedMessage, fetchAppInfo, fetchAppParams, removeMessage, saveMessage } from '@/service/share'
 import type { SiteInfo } from '@/models/share'
-import type { MoreLikeThisConfig, PromptConfig, SavedMessage } from '@/models/debug'
+import type {
+  MoreLikeThisConfig,
+  PromptConfig,
+  SavedMessage,
+  TextToSpeechConfig,
+} from '@/models/debug'
 import AppIcon from '@/app/components/base/app-icon'
 import { changeLanguage } from '@/i18n/i18next-config'
 import Loading from '@/app/components/base/loading'
@@ -74,6 +79,7 @@ const TextGeneration: FC<IMainProps> = ({
   const [canReplaceLogo, setCanReplaceLogo] = useState<boolean>(false)
   const [promptConfig, setPromptConfig] = useState<PromptConfig | null>(null)
   const [moreLikeThisConfig, setMoreLikeThisConfig] = useState<MoreLikeThisConfig | null>(null)
+  const [textToSpeechConfig, setTextToSpeechConfig] = useState<TextToSpeechConfig | null>(null)
 
   // save message
   const [savedMessages, setSavedMessages] = useState<SavedMessage[]>([])
@@ -231,7 +237,7 @@ const TextGeneration: FC<IMainProps> = ({
             return
           }
         }
-        if (varItem.required === false)
+        if (!varItem.required)
           return
 
         if (item[varIndex].trim() === '') {
@@ -347,7 +353,7 @@ const TextGeneration: FC<IMainProps> = ({
       setCanReplaceLogo(can_replace_logo)
       changeLanguage(siteInfo.default_language)
 
-      const { user_input_form, more_like_this, file_upload, sensitive_word_avoidance }: any = appParams
+      const { user_input_form, more_like_this, file_upload, text_to_speech, sensitive_word_avoidance }: any = appParams
       setVisionConfig({
         ...file_upload.image,
         image_file_size_limit: appParams?.system_parameters?.image_file_size_limit,
@@ -358,6 +364,7 @@ const TextGeneration: FC<IMainProps> = ({
         prompt_variables,
       } as PromptConfig)
       setMoreLikeThisConfig(more_like_this)
+      setTextToSpeechConfig(text_to_speech)
     })()
   }, [])
 
@@ -388,7 +395,7 @@ const TextGeneration: FC<IMainProps> = ({
     isCallBatchAPI={isCallBatchAPI}
     isPC={isPC}
     isMobile={isMobile}
-    isInstalledApp={!!isInstalledApp}
+    isInstalledApp={isInstalledApp}
     installedAppInfo={installedAppInfo}
     isError={task?.status === TaskStatus.failed}
     promptConfig={promptConfig}
@@ -403,6 +410,7 @@ const TextGeneration: FC<IMainProps> = ({
     onCompleted={handleCompleted}
     visionConfig={visionConfig}
     completionFiles={completionFiles}
+    isShowTextToSpeech={!!textToSpeechConfig?.enabled}
   />)
 
   const renderBatchRes = () => {
@@ -562,6 +570,7 @@ const TextGeneration: FC<IMainProps> = ({
             {currTab === 'saved' && (
               <SavedItems
                 className='mt-4'
+                isShowTextToSpeech={textToSpeechConfig?.enabled}
                 list={savedMessages}
                 onRemove={handleRemoveSavedMessage}
                 onStartCreateContent={() => setCurrTab('create')}

web/app/components/share/text-generation/result/index.tsx

@@ -22,6 +22,7 @@ export type IResultProps = {
   isInstalledApp: boolean
   installedAppInfo?: InstalledApp
   isError: boolean
+  isShowTextToSpeech: boolean
   promptConfig: PromptConfig | null
   moreLikeThisEnabled: boolean
   inputs: Record<string, any>
@@ -45,6 +46,7 @@ const Result: FC<IResultProps> = ({
   isInstalledApp,
   installedAppInfo,
   isError,
+  isShowTextToSpeech,
   promptConfig,
   moreLikeThisEnabled,
   inputs,
@@ -230,6 +232,7 @@ const Result: FC<IResultProps> = ({
       isLoading={isCallBatchAPI ? (!completionRes && isResponsing) : false}
       taskId={isCallBatchAPI ? ((taskId as number) < 10 ? `0${taskId}` : `${taskId}`) : undefined}
       controlClearMoreLikeThis={controlClearMoreLikeThis}
+      isShowTextToSpeech={isShowTextToSpeech}
     />
   )
 

web/context/debug-configuration.ts

@@ -16,6 +16,7 @@ import type {
   PromptItem,
   SpeechToTextConfig,
   SuggestedQuestionsAfterAnswerConfig,
+  TextToSpeechConfig,
 } from '@/models/debug'
 import type { ExternalDataTool } from '@/models/common'
 import type { DataSet } from '@/models/datasets'
@@ -64,6 +65,8 @@ type IDebugConfiguration = {
   setSuggestedQuestionsAfterAnswerConfig: (suggestedQuestionsAfterAnswerConfig: SuggestedQuestionsAfterAnswerConfig) => void
   speechToTextConfig: SpeechToTextConfig
   setSpeechToTextConfig: (speechToTextConfig: SpeechToTextConfig) => void
+  textToSpeechConfig: TextToSpeechConfig
+  setTextToSpeechConfig: (textToSpeechConfig: TextToSpeechConfig) => void
   citationConfig: CitationConfig
   setCitationConfig: (citationConfig: CitationConfig) => void
   annotationConfig: AnnotationReplyConfig
@@ -151,6 +154,10 @@ const DebugConfigurationContext = createContext<IDebugConfiguration>({
     enabled: false,
   },
   setSpeechToTextConfig: () => { },
+  textToSpeechConfig: {
+    enabled: false,
+  },
+  setTextToSpeechConfig: () => { },
   citationConfig: {
     enabled: false,
   },
@@ -197,6 +204,7 @@ const DebugConfigurationContext = createContext<IDebugConfiguration>({
     more_like_this: null,
     suggested_questions_after_answer: null,
     speech_to_text: null,
+    text_to_speech: null,
     retriever_resource: null,
     sensitive_word_avoidance: null,
     dataSets: [],

web/i18n/lang/app-api.en.ts

@@ -6,6 +6,9 @@ const translation = {
   ok: 'In Service',
   copy: 'Copy',
   copied: 'Copied',
+  play: 'Play',
+  pause: 'Pause',
+  playing: 'Playing',
   merMaind: {
     rerender: 'Redo Rerender',
   },

web/i18n/lang/app-api.zh.ts

@@ -5,10 +5,13 @@ const translation = {
   disabled: '已停用',
   ok: '运行中',
   copy: '复制',
+  copied: '已复制',
+  play: '播放',
+  pause: '暂停',
+  playing: '播放中',
   merMaind: {
     rerender: '重新渲染',
   },
-  copied: '已复制',
   never: '从未',
   apiKeyModal: {
     apiSecretKey: 'API 密钥',

web/i18n/lang/app-debug.en.ts

@@ -71,6 +71,11 @@ const translation = {
       description: 'Once enabled, you can use voice input.',
       resDes: 'Voice input is enabled',
     },
+    textToSpeech: {
+      title: 'Text to Speech',
+      description: 'Once enabled, text can be converted to speech.',
+      resDes: 'Text to Audio is enabled',
+    },
     citation: {
       title: 'Citations and Attributions',
       description: 'Once enabled, show source document and attributed section of the generated content.',

web/i18n/lang/app-debug.zh.ts

@@ -71,6 +71,11 @@ const translation = {
       description: '启用后，您可以使用语音输入。',
       resDes: '语音输入已启用',
     },
+    textToSpeech: {
+      title: '文字转语音',
+      description: '启用后，文本可以转换成语音。',
+      resDes: '文本转音频已启用',
+    },
     citation: {
       title: '引用和归属',
       description: '启用后，显示源文档和生成内容的归属部分。',

web/i18n/lang/common.en.ts

@@ -248,6 +248,10 @@ const translation = {
       key: 'Speech-to-Text Model',
       tip: 'Set the default model for speech-to-text input in conversation.',
     },
+    ttsModel: {
+      key: 'Speech-to-Text Model',
+      tip: 'Set the default model for speech-to-text input in conversation.',
+    },
     rerankModel: {
       key: 'Rerank Model',
       tip: 'Rerank model will reorder the candidate document list based on the semantic match with  user query, improving the results of semantic ranking',

web/i18n/lang/common.zh.ts

@@ -248,6 +248,10 @@ const translation = {
       key: '语音转文本模型',
       tip: '设置对话中语音转文字输入的默认使用模型。',
     },
+    ttsModel: {
+      key: '文本转语音模型',
+      tip: '设置对话中文字转语音输出的默认使用模型。',
+    },
     rerankModel: {
       key: 'Rerank 模型',
       tip: '重排序模型将根据候选文档列表与用户问题语义匹配度进行重新排序，从而改进语义排序的结果',

web/models/debug.ts

@@ -75,6 +75,8 @@ export type SuggestedQuestionsAfterAnswerConfig = MoreLikeThisConfig
 
 export type SpeechToTextConfig = MoreLikeThisConfig
 
+export type TextToSpeechConfig = MoreLikeThisConfig
+
 export type CitationConfig = MoreLikeThisConfig
 
 export type AnnotationReplyConfig = {
@@ -118,6 +120,7 @@ export type ModelConfig = {
   more_like_this: MoreLikeThisConfig | null
   suggested_questions_after_answer: SuggestedQuestionsAfterAnswerConfig | null
   speech_to_text: SpeechToTextConfig | null
+  text_to_speech: TextToSpeechConfig | null
   retriever_resource: RetrieverResourceConfig | null
   sensitive_word_avoidance: ModerationConfig | null
   dataSets: any[]

web/service/share.ts

@@ -136,6 +136,10 @@ export const audioToText = (url: string, isPublicAPI: boolean, body: FormData) =
   return (getAction('post', !isPublicAPI))(url, { body }, { bodyStringify: false, deleteContentType: true }) as Promise<{ text: string }>
 }
 
+export const textToAudio = (url: string, isPublicAPI: boolean, body: FormData) => {
+  return (getAction('post', !isPublicAPI))(url, { body }, { bodyStringify: false, deleteContentType: true }) as Promise<{ data: string }>
+}
+
 export const fetchAccessToken = async (appCode: string) => {
   const headers = new Headers()
   headers.append('X-App-Code', appCode)

web/types/app.ts

@@ -142,6 +142,9 @@ export type ModelConfig = {
   speech_to_text: {
     enabled: boolean
   }
+  text_to_speech: {
+    enabled: boolean
+  }
   retriever_resource: {
     enabled: boolean
   }